Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull second set of VFS changes from Al Viro:
 "Assorted f_pos race fixes, making do_splice_direct() safe to call with
  i_mutex on parent, O_TMPFILE support, Jeff's locks.c series,
  ->d_hash/->d_compare calling conventions changes from Linus, misc
  stuff all over the place."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
  Document ->tmpfile()
  ext4: ->tmpfile() support
  vfs: export lseek_execute() to modules
  lseek_execute() doesn't need an inode passed to it
  block_dev: switch to fixed_size_llseek()
  cpqphp_sysfs: switch to fixed_size_llseek()
  tile-srom: switch to fixed_size_llseek()
  proc_powerpc: switch to fixed_size_llseek()
  ubi/cdev: switch to fixed_size_llseek()
  pci/proc: switch to fixed_size_llseek()
  isapnp: switch to fixed_size_llseek()
  lpfc: switch to fixed_size_llseek()
  locks: give the blocked_hash its own spinlock
  locks: add a new "lm_owner_key" lock operation
  locks: turn the blocked_list into a hashtable
  locks: convert fl_link to a hlist_node
  locks: avoid taking global lock if possible when waking up blocked waiters
  locks: protect most of the file_lock handling with i_lock
  locks: encapsulate the fl_link list handling
  locks: make "added" in __posix_lock_file a bool
  ...

65 files changed, 800 insertions, 586 deletions

diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c
index ade28bb058e..0d138c0de29 100644
--- a/fs/adfs/dir.c
+++ b/fs/adfs/dir.c
@@ -191,8 +191,7 @@ const struct file_operations adfs_dir_operations = {
 };
 
 static int
-adfs_hash(const struct dentry *parent, const struct inode *inode,
-		struct qstr *qstr)
+adfs_hash(const struct dentry *parent, struct qstr *qstr)
 {
 	const unsigned int name_len = ADFS_SB(parent->d_sb)->s_namelen;
 	const unsigned char *name;
@@ -228,8 +227,7 @@ adfs_hash(const struct dentry *parent, const struct inode *inode,
  * requirements of the underlying filesystem.
  */
 static int
-adfs_compare(const struct dentry *parent, const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+adfs_compare(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name)
 {
 	int i;
diff --git a/fs/affs/namei.c b/fs/affs/namei.c
index ff65884a783..c36cbb4537a 100644
--- a/fs/affs/namei.c
+++ b/fs/affs/namei.c
@@ -13,18 +13,12 @@
 typedef int (*toupper_t)(int);
 
 static int	 affs_toupper(int ch);
-static int	 affs_hash_dentry(const struct dentry *,
-		const struct inode *, struct qstr *);
-static int       affs_compare_dentry(const struct dentry *parent,
-		const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+static int	 affs_hash_dentry(const struct dentry *, struct qstr *);
+static int       affs_compare_dentry(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name);
 static int	 affs_intl_toupper(int ch);
-static int	 affs_intl_hash_dentry(const struct dentry *,
-		const struct inode *, struct qstr *);
-static int       affs_intl_compare_dentry(const struct dentry *parent,
-		const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+static int	 affs_intl_hash_dentry(const struct dentry *, struct qstr *);
+static int       affs_intl_compare_dentry(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name);
 
 const struct dentry_operations affs_dentry_operations = {
@@ -86,14 +80,12 @@ __affs_hash_dentry(struct qstr *qstr, toupper_t toupper)
 }
 
 static int
-affs_hash_dentry(const struct dentry *dentry, const struct inode *inode,
-		struct qstr *qstr)
+affs_hash_dentry(const struct dentry *dentry, struct qstr *qstr)
 {
 	return __affs_hash_dentry(qstr, affs_toupper);
 }
 static int
-affs_intl_hash_dentry(const struct dentry *dentry, const struct inode *inode,
-		struct qstr *qstr)
+affs_intl_hash_dentry(const struct dentry *dentry, struct qstr *qstr)
 {
 	return __affs_hash_dentry(qstr, affs_intl_toupper);
 }
@@ -131,15 +123,13 @@ static inline int __affs_compare_dentry(unsigned int len,
 }
 
 static int
-affs_compare_dentry(const struct dentry *parent, const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+affs_compare_dentry(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name)
 {
 	return __affs_compare_dentry(len, str, name, affs_toupper);
 }
 static int
-affs_intl_compare_dentry(const struct dentry *parent,const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+affs_intl_compare_dentry(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name)
 {
 	return __affs_compare_dentry(len, str, name, affs_intl_toupper);
diff --git a/fs/afs/flock.c b/fs/afs/flock.c
index 2497bf306c7..a8cf2cff836 100644
--- a/fs/afs/flock.c
+++ b/fs/afs/flock.c
@@ -252,7 +252,8 @@ static void afs_defer_unlock(struct afs_vnode *vnode, struct key *key)
  */
 static int afs_do_setlk(struct file *file, struct file_lock *fl)
 {
-	struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host);
+	struct inode *inode = file_inode(file);
+	struct afs_vnode *vnode = AFS_FS_I(inode);
 	afs_lock_type_t type;
 	struct key *key = file->private_data;
 	int ret;
@@ -273,7 +274,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl)
 
 	type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE;
 
-	lock_flocks();
+	spin_lock(&inode->i_lock);
 
 	/* make sure we've got a callback on this file and that our view of the
 	 * data version is up to date */
@@ -420,7 +421,7 @@ given_lock:
 	afs_vnode_fetch_status(vnode, NULL, key);
 
 error:
-	unlock_flocks();
+	spin_unlock(&inode->i_lock);
 	_leave(" = %d", ret);
 	return ret;
 
diff --git a/fs/aio.c b/fs/aio.c
index 2bbcacf74d0..a8ecc8313fb 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -39,6 +39,8 @@
 #include <asm/kmap_types.h>
 #include <asm/uaccess.h>
 
+#include "internal.h"
+
 #define AIO_RING_MAGIC			0xa10a10a1
 #define AIO_RING_COMPAT_FEATURES	1
 #define AIO_RING_INCOMPAT_FEATURES	0
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 2091db8cdd7..431b6a04ebf 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -325,31 +325,10 @@ static int blkdev_write_end(struct file *file, struct address_space *mapping,
 static loff_t block_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct inode *bd_inode = file->f_mapping->host;
-	loff_t size;
 	loff_t retval;
 
 	mutex_lock(&bd_inode->i_mutex);
-	size = i_size_read(bd_inode);
-
-	retval = -EINVAL;
-	switch (whence) {
-		case SEEK_END:
-			offset += size;
-			break;
-		case SEEK_CUR:
-			offset += file->f_pos;
-		case SEEK_SET:
-			break;
-		default:
-			goto out;
-	}
-	if (offset >= 0 && offset <= size) {
-		if (offset != file->f_pos) {
-			file->f_pos = offset;
-		}
-		retval = offset;
-	}
-out:
+	retval = fixed_size_llseek(file, offset, whence, i_size_read(bd_inode));
 	mutex_unlock(&bd_inode->i_mutex);
 	return retval;
 }
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 4205ba752d4..89da56a58b6 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2425,20 +2425,7 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence)
 		}
 	}
 
-	if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) {
-		offset = -EINVAL;
-		goto out;
-	}
-	if (offset > inode->i_sb->s_maxbytes) {
-		offset = -EINVAL;
-		goto out;
-	}
-
-	/* Special lock needed here? */
-	if (offset != file->f_pos) {
-		file->f_pos = offset;
-		file->f_version = 0;
-	}
+	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 out:
 	mutex_unlock(&inode->i_mutex);
 	return offset;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 0f81d67cdc8..cd7e96c73cb 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -3881,7 +3881,7 @@ drop_write:
 
 static long btrfs_ioctl_quota_rescan(struct file *file, void __user *arg)
 {
-	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
 	struct btrfs_ioctl_quota_rescan_args *qsa;
 	int ret;
 
@@ -3914,7 +3914,7 @@ drop_write:
 
 static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg)
 {
-	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
 	struct btrfs_ioctl_quota_rescan_args *qsa;
 	int ret = 0;
 
@@ -4020,7 +4020,7 @@ out:
 
 static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg)
 {
-	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
 	const char *label = root->fs_info->super_copy->label;
 	size_t len = strnlen(label, BTRFS_LABEL_SIZE);
 	int ret;
@@ -4039,7 +4039,7 @@ static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg)
 
 static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg)
 {
-	struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root;
+	struct btrfs_root *root = BTRFS_I(file_inode(file))->root;
 	struct btrfs_super_block *super_block = root->fs_info->super_copy;
 	struct btrfs_trans_handle *trans;
 	char label[BTRFS_LABEL_SIZE];
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 656e1690743..16c989d3e23 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -866,16 +866,7 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int whence)
 		break;
 	}
 
-	if (offset < 0 || offset > inode->i_sb->s_maxbytes) {
-		offset = -EINVAL;
-		goto out;
-	}
-
-	/* Special lock needed here? */
-	if (offset != file->f_pos) {
-		file->f_pos = offset;
-		file->f_version = 0;
-	}
+	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 
 out:
 	mutex_unlock(&inode->i_mutex);
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index ebbf680378e..690f73f4242 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -192,7 +192,7 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count)
 
 /**
  * Encode the flock and fcntl locks for the given inode into the ceph_filelock
- * array. Must be called with lock_flocks() already held.
+ * array. Must be called with inode->i_lock already held.
  * If we encounter more of a specific lock type than expected, return -ENOSPC.
  */
 int ceph_encode_locks_to_buffer(struct inode *inode,
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 4d2920304be..74fd2898b2a 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -2481,20 +2481,20 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 		struct ceph_filelock *flocks;
 
 encode_again:
-		lock_flocks();
+		spin_lock(&inode->i_lock);
 		ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
-		unlock_flocks();
+		spin_unlock(&inode->i_lock);
 		flocks = kmalloc((num_fcntl_locks+num_flock_locks) *
 				 sizeof(struct ceph_filelock), GFP_NOFS);
 		if (!flocks) {
 			err = -ENOMEM;
 			goto out_free;
 		}
-		lock_flocks();
+		spin_lock(&inode->i_lock);
 		err = ceph_encode_locks_to_buffer(inode, flocks,
 						  num_fcntl_locks,
 						  num_flock_locks);
-		unlock_flocks();
+		spin_unlock(&inode->i_lock);
 		if (err) {
 			kfree(flocks);
 			if (err == -ENOSPC)
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 540c1ccfcdb..a445e71746f 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -765,7 +765,7 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int whence)
 
 static int cifs_setlease(struct file *file, long arg, struct file_lock **lease)
 {
-	/* note that this is called by vfs setlease with lock_flocks held
+	/* note that this is called by vfs setlease with i_lock held
 	   to protect *lease from going away */
 	struct inode *inode = file_inode(file);
 	struct cifsFileInfo *cfile = file->private_data;
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 5699b5036ed..5175aebf673 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -822,8 +822,7 @@ const struct dentry_operations cifs_dentry_ops = {
 /* d_delete:       cifs_d_delete,      */ /* not needed except for debugging */
 };
 
-static int cifs_ci_hash(const struct dentry *dentry, const struct inode *inode,
-		struct qstr *q)
+static int cifs_ci_hash(const struct dentry *dentry, struct qstr *q)
 {
 	struct nls_table *codepage = CIFS_SB(dentry->d_sb)->local_nls;
 	unsigned long hash;
@@ -838,12 +837,10 @@ static int cifs_ci_hash(const struct dentry *dentry, const struct inode *inode,
 	return 0;
 }
 
-static int cifs_ci_compare(const struct dentry *parent,
-		const struct inode *pinode,
-		const struct dentry *dentry, const struct inode *inode,
+static int cifs_ci_compare(const struct dentry *parent, const struct dentry *dentry,
 		unsigned int len, const char *str, const struct qstr *name)
 {
-	struct nls_table *codepage = CIFS_SB(pinode->i_sb)->local_nls;
+	struct nls_table *codepage = CIFS_SB(parent->d_sb)->local_nls;
 
 	if ((name->len == len) &&
 	    (nls_strnicmp(codepage, name->name, str, len) == 0))
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 4d8ba8d491e..91d8629e69a 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -999,7 +999,7 @@ try_again:
 		rc = wait_event_interruptible(flock->fl_wait, !flock->fl_next);
 		if (!rc)
 			goto try_again;
-		locks_delete_block(flock);
+		posix_unblock_lock(flock);
 	}
 	return rc;
 }
@@ -1092,6 +1092,7 @@ struct lock_to_push {
 static int
 cifs_push_posix_locks(struct cifsFileInfo *cfile)
 {
+	struct inode *inode = cfile->dentry->d_inode;
 	struct cifs_tcon *tcon = tlink_tcon(cfile->tlink);
 	struct file_lock *flock, **before;
 	unsigned int count = 0, i = 0;
@@ -1102,12 +1103,12 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 
 	xid = get_xid();
 
-	lock_flocks();
-	cifs_for_each_lock(cfile->dentry->d_inode, before) {
+	spin_lock(&inode->i_lock);
+	cifs_for_each_lock(inode, before) {
 		if ((*before)->fl_flags & FL_POSIX)
 			count++;
 	}
-	unlock_flocks();
+	spin_unlock(&inode->i_lock);
 
 	INIT_LIST_HEAD(&locks_to_send);
 
@@ -1126,8 +1127,8 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 	}
 
 	el = locks_to_send.next;
-	lock_flocks();
-	cifs_for_each_lock(cfile->dentry->d_inode, before) {
+	spin_lock(&inode->i_lock);
+	cifs_for_each_lock(inode, before) {
 		flock = *before;
 		if ((flock->fl_flags & FL_POSIX) == 0)
 			continue;
@@ -1152,7 +1153,7 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile)
 		lck->offset = flock->fl_start;
 		el = el->next;
 	}
-	unlock_flocks();
+	spin_unlock(&inode->i_lock);
 
 	list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) {
 		int stored_rc;
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 87e0ee9f446..14a14808320 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -487,13 +487,7 @@ static int coda_venus_readdir(struct file *coda_file, struct dir_context *ctx)
 
 		/* skip null entries */
 		if (vdir->d_fileno && name.len) {
-			/* try to look up this entry in the dcache, that way
-			 * userspace doesn't have to worry about breaking
-			 * getcwd by having mismatched inode numbers for
-			 * internal volume mountpoints. */
-			ino = find_inode_number(de, &name);
-			if (!ino) ino = vdir->d_fileno;
-
+			ino = vdir->d_fileno;
 			type = CDT2DT(vdir->d_type);
 			if (!dir_emit(ctx, name.name, name.len, ino, type))
 				break;
diff --git a/fs/dcache.c b/fs/dcache.c
index 5a23073138d..87bdb5329c3 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1730,7 +1730,7 @@ EXPORT_SYMBOL(d_add_ci);
  * Do the slow-case of the dentry name compare.
  *
  * Unlike the dentry_cmp() function, we need to atomically
- * load the name, length and inode information, so that the
+ * load the name and length information, so that the
  * filesystem can rely on them, and can use the 'name' and
  * 'len' information without worrying about walking off the
  * end of memory etc.
@@ -1748,22 +1748,18 @@ enum slow_d_compare {
 
 static noinline enum slow_d_compare slow_dentry_cmp(
 		const struct dentry *parent,
-		struct inode *inode,
 		struct dentry *dentry,
 		unsigned int seq,
 		const struct qstr *name)
 {
 	int tlen = dentry->d_name.len;
 	const char *tname = dentry->d_name.name;
-	struct inode *i = dentry->d_inode;
 
 	if (read_seqcount_retry(&dentry->d_seq, seq)) {
 		cpu_relax();
 		return D_COMP_SEQRETRY;
 	}
-	if (parent->d_op->d_compare(parent, inode,
-				dentry, i,
-				tlen, tname, name))
+	if (parent->d_op->d_compare(parent, dentry, tlen, tname, name))
 		return D_COMP_NOMATCH;
 	return D_COMP_OK;
 }
@@ -1773,7 +1769,6 @@ static noinline enum slow_d_compare slow_dentry_cmp(
  * @parent: parent dentry
  * @name: qstr of name we wish to find
  * @seqp: returns d_seq value at the point where the dentry was found
- * @inode: returns dentry->d_inode when the inode was found valid.
  * Returns: dentry, or NULL
  *
  * __d_lookup_rcu is the dcache lookup function for rcu-walk name
@@ -1800,7 +1795,7 @@ static noinline enum slow_d_compare slow_dentry_cmp(
  */
 struct dentry *__d_lookup_rcu(const struct dentry *parent,
 				const struct qstr *name,
-				unsigned *seqp, struct inode *inode)
+				unsigned *seqp)
 {
 	u64 hashlen = name->hash_len;
 	const unsigned char *str = name->name;
@@ -1834,11 +1829,10 @@ struct dentry *__d_lookup_rcu(const struct dentry *parent,
 seqretry:
 		/*
 		 * The dentry sequence count protects us from concurrent
-		 * renames, and thus protects inode, parent and name fields.
+		 * renames, and thus protects parent and name fields.
 		 *
 		 * The caller must perform a seqcount check in order
-		 * to do anything useful with the returned dentry,
-		 * including using the 'd_inode' pointer.
+		 * to do anything useful with the returned dentry.
 		 *
 		 * NOTE! We do a "raw" seqcount_begin here. That means that
 		 * we don't wait for the sequence count to stabilize if it
@@ -1852,12 +1846,12 @@ seqretry:
 			continue;
 		if (d_unhashed(dentry))
 			continue;
-		*seqp = seq;
 
 		if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) {
 			if (dentry->d_name.hash != hashlen_hash(hashlen))
 				continue;
-			switch (slow_dentry_cmp(parent, inode, dentry, seq, name)) {
+			*seqp = seq;
+			switch (slow_dentry_cmp(parent, dentry, seq, name)) {
 			case D_COMP_OK:
 				return dentry;
 			case D_COMP_NOMATCH:
@@ -1869,6 +1863,7 @@ seqretry:
 
 		if (dentry->d_name.hash_len != hashlen)
 			continue;
+		*seqp = seq;
 		if (!dentry_cmp(dentry, str, hashlen_len(hashlen)))
 			return dentry;
 	}
@@ -1966,9 +1961,7 @@ struct dentry *__d_lookup(const struct dentry *parent, const struct qstr *name)
 		if (parent->d_flags & DCACHE_OP_COMPARE) {
 			int tlen = dentry->d_name.len;
 			const char *tname = dentry->d_name.name;
-			if (parent->d_op->d_compare(parent, parent->d_inode,
-						dentry, dentry->d_inode,
-						tlen, tname, name))
+			if (parent->d_op->d_compare(parent, dentry, tlen, tname, name))
 				goto next;
 		} else {
 			if (dentry->d_name.len != len)
@@ -2005,7 +1998,7 @@ struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name)
 	 */
 	name->hash = full_name_hash(name->name, name->len);