Merge branch 'master' of /pub/scm/linux/kernel/git/torvalds/linux-2.6

66 files changed, 1047 insertions, 1009 deletions

diff --git a/fs/autofs/dirhash.c b/fs/autofs/dirhash.c
index bf8c8af9800..4eb4d8dfb2f 100644
--- a/fs/autofs/dirhash.c
+++ b/fs/autofs/dirhash.c
@@ -39,10 +39,12 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 {
 	struct autofs_dirhash *dh = &sbi->dirhash;
 	struct autofs_dir_ent *ent;
-	struct dentry *dentry;
 	unsigned long timeout = sbi->exp_timeout;
 
 	while (1) {
+		struct path path;
+		int umount_ok;
+
 		if ( list_empty(&dh->expiry_head) || sbi->catatonic )
 			return NULL;	/* No entries */
 		/* We keep the list sorted by last_usage and want old stuff */
@@ -57,17 +59,17 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 			return ent; /* Symlinks are always expirable */
 
 		/* Get the dentry for the autofs subdirectory */
-		dentry = ent->dentry;
+		path.dentry = ent->dentry;
 
-		if ( !dentry ) {
+		if (!path.dentry) {
 			/* Should only happen in catatonic mode */
 			printk("autofs: dentry == NULL but inode range is directory, entry %s\n", ent->name);
 			autofs_delete_usage(ent);
 			continue;
 		}
 
-		if ( !dentry->d_inode ) {
-			dput(dentry);
+		if (!path.dentry->d_inode) {
+			dput(path.dentry);
 			printk("autofs: negative dentry on expiry queue: %s\n",
 			       ent->name);
 			autofs_delete_usage(ent);
@@ -76,29 +78,29 @@ struct autofs_dir_ent *autofs_expire(struct super_block *sb,
 
 		/* Make sure entry is mounted and unused; note that dentry will
 		   point to the mounted-on-top root. */
-		if (!S_ISDIR(dentry->d_inode->i_mode)||!d_mountpoint(dentry)) {
+		if (!S_ISDIR(path.dentry->d_inode->i_mode) ||
+		    !d_mountpoint(path.dentry)) {
 			DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name));
 			continue;
 		}
-		mntget(mnt);
-		dget(dentry);
-		if (!follow_down(&mnt, &dentry)) {
-			dput(dentry);
-			mntput(mnt);
+		path.mnt = mnt;
+		path_get(&path);
+		if (!follow_down(&path.mnt, &path.dentry)) {
+			path_put(&path);
 			DPRINTK(("autofs: not expirable (not a mounted directory): %s\n", ent->name));
 			continue;
 		}
-		while (d_mountpoint(dentry) && follow_down(&mnt, &dentry))
+		while (d_mountpoint(path.dentry) &&
+		       follow_down(&path.mnt, &path.dentry))
 			;
-		dput(dentry);
+		umount_ok = may_umount(path.mnt);
+		path_put(&path);
 
-		if ( may_umount(mnt) ) {
-			mntput(mnt);
+		if (umount_ok) {
 			DPRINTK(("autofs: signaling expire on %s\n", ent->name));
 			return ent; /* Expirable! */
 		}
 		DPRINTK(("autofs: didn't expire due to may_umount: %s\n", ent->name));
-		mntput(mnt);
 	}
 	return NULL;		/* No expirable entries */
 }
diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index 9e5ae8a4f5c..84168c0dcc2 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -54,11 +54,10 @@ static int check_name(const char *name)
  * Check a string doesn't overrun the chunk of
  * memory we copied from user land.
  */
-static int invalid_str(char *str, void *end)
+static int invalid_str(char *str, size_t size)
 {
-	while ((void *) str <= end)
-		if (!*str++)
-			return 0;
+	if (memchr(str, 0, size))
+		return 0;
 	return -EINVAL;
 }
 
@@ -138,8 +137,7 @@ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param)
 	}
 
 	if (param->size > sizeof(*param)) {
-		err = invalid_str(param->path,
-				 (void *) ((size_t) param + param->size));
+		err = invalid_str(param->path, param->size - sizeof(*param));
 		if (err) {
 			AUTOFS_WARN(
 			  "path string terminator missing for cmd(0x%08x)",
@@ -488,7 +486,7 @@ static int autofs_dev_ioctl_requester(struct file *fp,
 	}
 
 	path = param->path;
-	devid = sbi->sb->s_dev;
+	devid = new_encode_dev(sbi->sb->s_dev);
 
 	param->requester.uid = param->requester.gid = -1;
 
diff --git a/fs/bio.c b/fs/bio.c
index cd42bb882f3..98711647ece 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -175,14 +175,6 @@ struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx,
 	struct bio_vec *bvl;
 
 	/*
-	 * If 'bs' is given, lookup the pool and do the mempool alloc.
-	 * If not, this is a bio_kmalloc() allocation and just do a
-	 * kzalloc() for the exact number of vecs right away.
-	 */
-	if (!bs)
-		bvl = kmalloc(nr * sizeof(struct bio_vec), gfp_mask);
-
-	/*
 	 * see comment near bvec_array define!
 	 */
 	switch (nr) {
@@ -260,21 +252,6 @@ void bio_free(struct bio *bio, struct bio_set *bs)
 	mempool_free(p, bs->bio_pool);
 }
 
-/*
- * default destructor for a bio allocated with bio_alloc_bioset()
- */
-static void bio_fs_destructor(struct bio *bio)
-{
-	bio_free(bio, fs_bio_set);
-}
-
-static void bio_kmalloc_destructor(struct bio *bio)
-{
-	if (bio_has_allocated_vec(bio))
-		kfree(bio->bi_io_vec);
-	kfree(bio);
-}
-
 void bio_init(struct bio *bio)
 {
 	memset(bio, 0, sizeof(*bio));
@@ -301,21 +278,15 @@ void bio_init(struct bio *bio)
  **/
 struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs)
 {
+	unsigned long idx = BIO_POOL_NONE;
 	struct bio_vec *bvl = NULL;
-	struct bio *bio = NULL;
-	unsigned long idx = 0;
-	void *p = NULL;
-
-	if (bs) {
-		p = mempool_alloc(bs->bio_pool, gfp_mask);
-		if (!p)
-			goto err;
-		bio = p + bs->front_pad;
-	} else {
-		bio = kmalloc(sizeof(*bio), gfp_mask);
-		if (!bio)
-			goto err;
-	}
+	struct bio *bio;
+	void *p;
+
+	p = mempool_alloc(bs->bio_pool, gfp_mask);
+	if (unlikely(!p))
+		return NULL;
+	bio = p + bs->front_pad;
 
 	bio_init(bio);
 
@@ -332,22 +303,50 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs)
 
 		nr_iovecs = bvec_nr_vecs(idx);
 	}
+out_set:
 	bio->bi_flags |= idx << BIO_POOL_OFFSET;
 	bio->bi_max_vecs = nr_iovecs;
-out_set:
 	bio->bi_io_vec = bvl;
-
 	return bio;
 
 err_free:
-	if (bs)
-		mempool_free(p, bs->bio_pool);
-	else
-		kfree(bio);
-err:
+	mempool_free(p, bs->bio_pool);
 	return NULL;
 }
 
+static void bio_fs_destructor(struct bio *bio)
+{
+	bio_free(bio, fs_bio_set);
+}
+
+/**
+ *	bio_alloc - allocate a new bio, memory pool backed
+ *	@gfp_mask: allocation mask to use
+ *	@nr_iovecs: number of iovecs
+ *
+ *	Allocate a new bio with @nr_iovecs bvecs.  If @gfp_mask
+ *	contains __GFP_WAIT, the allocation is guaranteed to succeed.
+ *
+ *	RETURNS:
+ *	Pointer to new bio on success, NULL on failure.
+ */
+struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs)
+{
+	struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set);
+
+	if (bio)
+		bio->bi_destructor = bio_fs_destructor;
+
+	return bio;
+}
+
+static void bio_kmalloc_destructor(struct bio *bio)
+{
+	if (bio_integrity(bio))
+		bio_integrity_free(bio);
+	kfree(bio);
+}
+
 /**
  * bio_alloc - allocate a bio for I/O
  * @gfp_mask:   the GFP_ mask given to the slab allocator
@@ -366,29 +365,20 @@ err:
  *   do so can cause livelocks under memory pressure.
  *
  **/
-struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs)
-{
-	struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set);
-
-	if (bio)
-		bio->bi_destructor = bio_fs_destructor;
-
-	return bio;
-}
-
-/*
- * Like bio_alloc(), but doesn't use a mempool backing. This means that
- * it CAN fail, but while bio_alloc() can only be used for allocations
- * that have a short (finite) life span, bio_kmalloc() should be used
- * for more permanent bio allocations (like allocating some bio's for
- * initalization or setup purposes).
- */
 struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs)
 {
-	struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, NULL);
+	struct bio *bio;
 
-	if (bio)
-		bio->bi_destructor = bio_kmalloc_destructor;
+	bio = kmalloc(sizeof(struct bio) + nr_iovecs * sizeof(struct bio_vec),
+		      gfp_mask);
+	if (unlikely(!bio))
+		return NULL;
+
+	bio_init(bio);
+	bio->bi_flags |= BIO_POOL_NONE << BIO_POOL_OFFSET;
+	bio->bi_max_vecs = nr_iovecs;
+	bio->bi_io_vec = bio->bi_inline_vecs;
+	bio->bi_destructor = bio_kmalloc_destructor;
 
 	return bio;
 }
@@ -827,12 +817,15 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
 		len += iov[i].iov_len;
 	}
 
+	if (offset)
+		nr_pages++;
+
 	bmd = bio_alloc_map_data(nr_pages, iov_count, gfp_mask);
 	if (!bmd)
 		return ERR_PTR(-ENOMEM);
 
 	ret = -ENOMEM;
-	bio = bio_alloc(gfp_mask, nr_pages);
+	bio = bio_kmalloc(gfp_mask, nr_pages);
 	if (!bio)
 		goto out_bmd;
 
@@ -956,7 +949,7 @@ static struct bio *__bio_map_user_iov(struct request_queue *q,
 	if (!nr_pages)
 		return ERR_PTR(-EINVAL);
 
-	bio = bio_alloc(gfp_mask, nr_pages);
+	bio = bio_kmalloc(gfp_mask, nr_pages);
 	if (!bio)
 		return ERR_PTR(-ENOMEM);
 
@@ -1140,7 +1133,7 @@ static struct bio *__bio_map_kern(struct request_queue *q, void *data,
 	int offset, i;
 	struct bio *bio;
 
-	bio = bio_alloc(gfp_mask, nr_pages);
+	bio = bio_kmalloc(gfp_mask, nr_pages);
 	if (!bio)
 		return ERR_PTR(-ENOMEM);
 
diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile
index 9adf5e4f7e9..94212844a9b 100644
--- a/fs/btrfs/Makefile
+++ b/fs/btrfs/Makefile
@@ -1,25 +1,10 @@
-ifneq ($(KERNELRELEASE),)
-# kbuild part of makefile
 
 obj-$(CONFIG_BTRFS_FS) := btrfs.o
-btrfs-y := super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
+
+btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \
 	   file-item.o inode-item.o inode-map.o disk-io.o \
 	   transaction.o inode.o file.o tree-defrag.o \
 	   extent_map.o sysfs.o struct-funcs.o xattr.o ordered-data.o \
 	   extent_io.o volumes.o async-thread.o ioctl.o locking.o orphan.o \
 	   ref-cache.o export.o tree-log.o acl.o free-space-cache.o zlib.o \
 	   compression.o delayed-ref.o
-else
-
-# Normal Makefile
-
-KERNELDIR := /lib/modules/`uname -r`/build
-all:
-	$(MAKE) -C $(KERNELDIR) M=`pwd` CONFIG_BTRFS_FS=m modules
-
-modules_install:
-	$(MAKE) -C $(KERNELDIR) M=`pwd` modules_install
-clean:
-	$(MAKE) -C $(KERNELDIR) M=`pwd` clean
-
-endif
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 7fdd184a528..cbba000dccb 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -60,15 +60,20 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 		return ERR_PTR(-EINVAL);
 	}
 
+	/* Handle the cached NULL acl case without locking */
+	acl = ACCESS_ONCE(*p_acl);
+	if (!acl)
+		return acl;
+
 	spin_lock(&inode->i_lock);
-	if (*p_acl != BTRFS_ACL_NOT_CACHED)
-		acl = posix_acl_dup(*p_acl);
+	acl = *p_acl;
+	if (acl != BTRFS_ACL_NOT_CACHED)
+		acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
 
-	if (acl)
+	if (acl != BTRFS_ACL_NOT_CACHED)
 		return acl;
 
-
 	size = __btrfs_getxattr(inode, name, "", 0);
 	if (size > 0) {
 		value = kzalloc(size, GFP_NOFS);
@@ -80,9 +85,12 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 			btrfs_update_cached_acl(inode, p_acl, acl);
 		}
 		kfree(value);
-	} else if (size == -ENOENT) {
+	} else if (size == -ENOENT || size == -ENODATA || size == 0) {
+		/* FIXME, who returns -ENOENT?  I think nobody */
 		acl = NULL;
 		btrfs_update_cached_acl(inode, p_acl, acl);
+	} else {
+		acl = ERR_PTR(-EIO);
 	}
 
 	return acl;
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 51bfdfc8fcd..502c3d61de6 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -25,6 +25,7 @@
 #define WORK_QUEUED_BIT 0
 #define WORK_DONE_BIT 1
 #define WORK_ORDER_DONE_BIT 2
+#define WORK_HIGH_PRIO_BIT 3
 
 /*
  * container for the kthread task pointer and the list of pending work
@@ -36,6 +37,7 @@ struct btrfs_worker_thread {
 
 	/* list of struct btrfs_work that are waiting for service */
 	struct list_head pending;
+	struct list_head prio_pending;
 
 	/* list of worker threads from struct btrfs_workers */
 	struct list_head worker_list;
@@ -103,10 +105,16 @@ static noinline int run_ordered_completions(struct btrfs_workers *workers,
 
 	spin_lock_irqsave(&workers->lock, flags);
 
-	while (!list_empty(&workers->order_list)) {
-		work = list_entry(workers->order_list.next,
-				  struct btrfs_work, order_list);
-
+	while (1) {
+		if (!list_empty(&workers->prio_order_list)) {
+			work = list_entry(workers->prio_order_list.next,
+					  struct btrfs_work, order_list);
+		} else if (!list_empty(&workers->order_list)) {
+			work = list_entry(workers->order_list.next,
+					  struct btrfs_work, order_list);
+		} else {
+			break;
+		}
 		if (!test_bit(WORK_DONE_BIT, &work->flags))
 			break;
 
@@ -143,8 +151,14 @@ static int worker_loop(void *arg)
 	do {
 		spin_lock_irq(&worker->lock);
 again_locked:
-		while (!list_empty(&worker->pending)) {
-			cur = worker->pending.next;
+		while (1) {
+			if (!list_empty(&worker->prio_pending))
+				cur = worker->prio_pending.next;
+			else if (!list_empty(&worker->pending))
+				cur = worker->pending.next;
+			else
+				break;
+
 			work = list_entry(cur, struct btrfs_work, list);
 			list_del(&work->list);
 			clear_bit(WORK_QUEUED_BIT, &work->flags);
@@ -163,7 +177,6 @@ again_locked:
 
 			spin_lock_irq(&worker->lock);
 			check_idle_worker(worker);
-
 		}
 		if (freezing(current)) {
 			worker->working = 0;
@@ -178,7 +191,8 @@ again_locked:
 				 * jump_in?
 				 */
 				smp_mb();
-				if (!list_empty(&worker->pending))
+				if (!list_empty(&worker->pending) ||
+				    !list_empty(&worker->prio_pending))
 					continue;
 
 				/*
@@ -191,7 +205,8 @@ again_locked:
 				 */
 				schedule_timeout(1);
 				smp_mb();
-				if (!list_empty(&worker->pending))
+				if (!list_empty(&worker->pending) ||
+				    !list_empty(&worker->prio_pending))
 					continue;
 
 				if (kthread_should_stop())
@@ -200,7 +215,8 @@ again_locked:
 				/* still no more work?, sleep for real */
 				spin_lock_irq(&worker->lock);
 				set_current_state(TASK_INTERRUPTIBLE);
-				if (!list_empty(&worker->pending))
+				if (!list_empty(&worker->pending) ||
+				    !list_empty(&worker->prio_pending))
 					goto again_locked;
 
 				/*
@@ -248,6 +264,7 @@ void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max)
 	INIT_LIST_HEAD(&workers->worker_list);
 	INIT_LIST_HEAD(&workers->idle_list);
 	INIT_LIST_HEAD(&workers->order_list);
+	INIT_LIST_HEAD(&workers->prio_order_list);
 	spin_lock_init(&workers->lock);
 	workers->max_workers = max;
 	workers->idle_thresh = 32;
@@ -273,6 +290,7 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
 		}
 
 		INIT_LIST_HEAD(&worker->pending);
+		INIT_LIST_HEAD(&worker->prio_pending);
 		INIT_LIST_HEAD(&worker->worker_list);
 		spin_lock_init(&worker->lock);
 		atomic_set(&worker->num_pending, 0);
@@ -396,7 +414,10 @@ int btrfs_requeue_work(struct btrfs_work *work)
 		goto out;
 
 	spin_lock_irqsave(&worker->lock, flags);
-	list_add_tail(&work->list, &worker->pending);
+	if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags))
+		list_add_tail(&work->list, &worker->prio_pending);
+	else
+		list_add_tail(&work->list, &worker->pending);
 	atomic_inc(&worker->num_pending);
 
 	/* by definition we're busy, take ourselves off the idle
@@ -422,6 +443,11 @@ out:
 	return 0;
 }
 
+void btrfs_set_work_high_prio(struct btrfs_work *work)
+{
+	set_bit(WORK_HIGH_PRIO_BIT, &work->flags);
+}
+
 /*
  * places a struct btrfs_work into the pending queue of one of the kthreads
  */
@@ -438,7 +464,12 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work)
 	worker = find_worker(workers);
 	if (workers->ordered) {
 		spin_lock_irqsave(&workers->lock, flags);
-		list_add_tail(&work->order_list, &workers->order_list);
+		if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags)) {
+			list_add_tail(&work->order_list,
+				      &workers->prio_order_list);
+		} else {
+			list_add_tail(&work->order_list, &workers->order_list);
+		}
 		spin_unlock_irqrestore(&workers->lock, flags);
 	} else {
 		INIT_LIST_HEAD(&work->order_list);
@@ -446,7 +477,10 @@ int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work)
 
 	spin_lock_irqsave(&worker->lock, flags);
 
-	list_add_tail(&work->list, &worker->pending);
+	if (test_bit(WORK_HIGH_PRIO_BIT, &work->flags))
+		list_add_tail(&work->list, &worker->prio_pending);
+	else
+		list_add_tail(&work->list, &worker->pending);
 	atomic_inc(&worker->num_pending);
 	check_busy_worker(worker);
 
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index 31be4ed8b63..1b511c109db 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -85,6 +85,7 @@ struct btrfs_workers {
 	 * of work items waiting for completion
 	 */
 	struct list_head order_list;
+	struct list_head prio_order_list;
 
 	/* lock for finding the next worker thread to queue on */
 	spinlock_t lock;
@@ -98,4 +99,5 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers);
 int btrfs_stop_workers(struct btrfs_workers *workers);
 void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max);
 int btrfs_requeue_work(struct btrfs_work *work);
+void btrfs_set_work_high_prio(struct btrfs_work *work);
 #endif
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index e5b2533b691..a99f1c2a710 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1325,12 +1325,12 @@ static noinline int reada_for_balance(struct btrfs_root *root,
 	int ret = 0;
 	int blocksize;
 
-	parent = path->nodes[level - 1];
+	parent = path->nodes[level + 1];
 	if (!parent)
 		return 0;
 
 	nritems = btrfs_header_nritems(parent);
-	slot = path->slots[level];
+	slot = path->slots[level + 1];
 	blocksize = btrfs_level_size(root, level);
 
 	if (slot > 0) {
@@ -1341,7 +1341,7 @@ static noinline int reada_for_balance(struct btrfs_root *root,
 			block1 = 0;
 		free_extent_buffer(eb);
 	}