Merge branch 'work.misc-set_fs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

Pull misc user access cleanups from Al Viro:
 "The first pile is assorted getting rid of cargo-culted access_ok(),
  cargo-culted set_fs() and field-by-field copyouts.

  The same description applies to a lot of stuff in other branches -
  this is just the stuff that didn't fit into a more specific topical
  branch"

* 'work.misc-set_fs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  Switch flock copyin/copyout primitives to copy_{from,to}_user()
  fs/fcntl: return -ESRCH in f_setown when pid/pgid can't be found
  fs/fcntl: f_setown, avoid undefined behaviour
  fs/fcntl: f_setown, allow returning error
  lpfc debugfs: get rid of pointless access_ok()
  adb: get rid of pointless access_ok()
  isdn: get rid of pointless access_ok()
  compat statfs: switch to copy_to_user()
  fs/locks: don't mess with the address limit in compat_fcntl64
  nfsd_readlink(): switch to vfs_get_link()
  drbd: ->sendpage() never needed set_fs()
  fs/locks: pass kernel struct flock to fcntl_getlk/setlk
  fs: locks: Fix some troubles at kernel-doc comments

drivers/block/drbd/drbd_main.c

@@ -1551,7 +1551,6 @@ static int _drbd_send_page(struct drbd_peer_device *peer_device, struct page *pa
 		    int offset, size_t size, unsigned msg_flags)
 {
 	struct socket *socket = peer_device->connection->data.socket;
-	mm_segment_t oldfs = get_fs();
 	int len = size;
 	int err = -EIO;
 
@@ -1566,7 +1565,6 @@ static int _drbd_send_page(struct drbd_peer_device *peer_device, struct page *pa
 
 	msg_flags |= MSG_NOSIGNAL;
 	drbd_update_congested(peer_device->connection);
-	set_fs(KERNEL_DS);
 	do {
 		int sent;
 
@@ -1586,7 +1584,6 @@ static int _drbd_send_page(struct drbd_peer_device *peer_device, struct page *pa
 		len    -= sent;
 		offset += sent;
 	} while (len > 0 /* THINK && device->cstate >= C_CONNECTED*/);
-	set_fs(oldfs);
 	clear_bit(NET_CONGESTED, &peer_device->connection->flags);
 
 	if (len == 0) {

drivers/isdn/i4l/isdn_common.c

@@ -1304,9 +1304,6 @@ isdn_ioctl(struct file *file, uint cmd, ulong arg)
 			if (arg) {
 				ulong __user *p = argp;
 				int i;
-				if (!access_ok(VERIFY_WRITE, p,
-					       sizeof(ulong) * ISDN_MAX_CHANNELS * 2))
-					return -EFAULT;
 				for (i = 0; i < ISDN_MAX_CHANNELS; i++) {
 					put_user(dev->ibytes[i], p++);
 					put_user(dev->obytes[i], p++);
@@ -1540,11 +1537,6 @@ isdn_ioctl(struct file *file, uint cmd, ulong arg)
 				char __user *p = argp;
 				int i;
 
-				if (!access_ok(VERIFY_WRITE, argp,
-					       (ISDN_MODEM_NUMREG + ISDN_MSNLEN + ISDN_LMSNLEN)
-					       * ISDN_MAX_CHANNELS))
-					return -EFAULT;
-
 				for (i = 0; i < ISDN_MAX_CHANNELS; i++) {
 					if (copy_to_user(p, dev->mdm.info[i].emu.profile,
 							 ISDN_MODEM_NUMREG))
@@ -1567,11 +1559,6 @@ isdn_ioctl(struct file *file, uint cmd, ulong arg)
 				char __user *p = argp;
 				int i;
 
-				if (!access_ok(VERIFY_READ, argp,
-					       (ISDN_MODEM_NUMREG + ISDN_MSNLEN + ISDN_LMSNLEN)
-					       * ISDN_MAX_CHANNELS))
-					return -EFAULT;
-
 				for (i = 0; i < ISDN_MAX_CHANNELS; i++) {
 					if (copy_from_user(dev->mdm.info[i].emu.profile, p,
 							   ISDN_MODEM_NUMREG))
@@ -1617,8 +1604,6 @@ isdn_ioctl(struct file *file, uint cmd, ulong arg)
 						int j = 0;
 
 						while (1) {
-							if (!access_ok(VERIFY_READ, p, 1))
-								return -EFAULT;
 							get_user(bname[j], p++);
 							switch (bname[j]) {
 							case '\0':
@@ -1685,9 +1670,6 @@ isdn_ioctl(struct file *file, uint cmd, ulong arg)
 					drvidx = 0;
 				if (drvidx == -1)
 					return -ENODEV;
-				if (!access_ok(VERIFY_WRITE, argp,
-					       sizeof(isdn_ioctl_struct)))
-					return -EFAULT;
 				c.driver = drvidx;
 				c.command = ISDN_CMD_IOCTL;
 				c.arg = cmd;

drivers/isdn/i4l/isdn_ppp.c

@@ -795,9 +795,6 @@ isdn_ppp_read(int min, struct file *file, char __user *buf, int count)
 	if (!(is->state & IPPP_OPEN))
 		return 0;
 
-	if (!access_ok(VERIFY_WRITE, buf, count))
-		return -EFAULT;
-
 	spin_lock_irqsave(&is->buflock, flags);
 	b = is->first->next;
 	save_buf = b->buf;
@@ -2014,9 +2011,6 @@ isdn_ppp_dev_ioctl_stats(int slot, struct ifreq *ifr, struct net_device *dev)
 	struct ppp_stats t;
 	isdn_net_local *lp = netdev_priv(dev);
 
-	if (!access_ok(VERIFY_WRITE, res, sizeof(struct ppp_stats)))
-		return -EFAULT;
-
 	/* build a temporary stat struct and copy it to user space */
 
 	memset(&t, 0, sizeof(struct ppp_stats));

drivers/isdn/isdnloop/isdnloop.c

@@ -1142,8 +1142,6 @@ isdnloop_command(isdn_ctrl *c, isdnloop_card *card)
 		case ISDNLOOP_IOCTL_DEBUGVAR:
 			return (ulong) card;
 		case ISDNLOOP_IOCTL_STARTUP:
-			if (!access_ok(VERIFY_READ, (void *) a, sizeof(isdnloop_sdef)))
-				return -EFAULT;
 			return isdnloop_start(card, (isdnloop_sdef *) a);
 			break;
 		case ISDNLOOP_IOCTL_ADDCARD:

drivers/macintosh/adb.c

@@ -723,8 +723,6 @@ static ssize_t adb_read(struct file *file, char __user *buf,
 		return -EINVAL;
 	if (count > sizeof(req->reply))
 		count = sizeof(req->reply);
-	if (!access_ok(VERIFY_WRITE, buf, count))
-		return -EFAULT;
 
 	req = NULL;
 	spin_lock_irqsave(&state->lock, flags);
@@ -781,8 +779,6 @@ static ssize_t adb_write(struct file *file, const char __user *buf,
 		return -EINVAL;
 	if (adb_controller == NULL)
 		return -ENXIO;
-	if (!access_ok(VERIFY_READ, buf, count))
-		return -EFAULT;
 
 	req = kmalloc(sizeof(struct adb_request),
 					     GFP_KERNEL);

drivers/scsi/lpfc/lpfc_debugfs.c

@@ -1949,10 +1949,6 @@ lpfc_debugfs_nvmestat_write(struct file *file, const char __user *buf,
 	if (nbytes > 64)
 		nbytes = 64;
 
-	/* Protect copy from user */
-	if (!access_ok(VERIFY_READ, buf, nbytes))
-		return -EFAULT;
-
 	memset(mybuf, 0, sizeof(mybuf));
 
 	if (copy_from_user(mybuf, buf, nbytes))
@@ -2037,10 +2033,6 @@ lpfc_debugfs_nvmektime_write(struct file *file, const char __user *buf,
 	if (nbytes > 64)
 		nbytes = 64;
 
-	/* Protect copy from user */
-	if (!access_ok(VERIFY_READ, buf, nbytes))
-		return -EFAULT;
-
 	memset(mybuf, 0, sizeof(mybuf));
 
 	if (copy_from_user(mybuf, buf, nbytes))
@@ -2169,10 +2161,6 @@ lpfc_debugfs_nvmeio_trc_write(struct file *file, const char __user *buf,
 	if (nbytes > 64)
 		nbytes = 64;
 
-	/* Protect copy from user */
-	if (!access_ok(VERIFY_READ, buf, nbytes))
-		return -EFAULT;
-
 	memset(mybuf, 0, sizeof(mybuf));
 
 	if (copy_from_user(mybuf, buf, nbytes))
@@ -2280,10 +2268,6 @@ lpfc_debugfs_cpucheck_write(struct file *file, const char __user *buf,
 	if (nbytes > 64)
 		nbytes = 64;
 
-	/* Protect copy from user */
-	if (!access_ok(VERIFY_READ, buf, nbytes))
-		return -EFAULT;
-
 	memset(mybuf, 0, sizeof(mybuf));
 
 	if (copy_from_user(mybuf, buf, nbytes))
@@ -2354,10 +2338,6 @@ static int lpfc_idiag_cmd_get(const char __user *buf, size_t nbytes,
 	int i;
 	size_t bsize;
 
-	/* Protect copy from user */
-	if (!access_ok(VERIFY_READ, buf, nbytes))
-		return -EFAULT;
-
 	memset(mybuf, 0, sizeof(mybuf));
 	memset(idiag_cmd, 0, sizeof(*idiag_cmd));
 	bsize = min(nbytes, (sizeof(mybuf)-1));

fs/fcntl.c

@@ -109,20 +109,34 @@ void __f_setown(struct file *filp, struct pid *pid, enum pid_type type,
 }
 EXPORT_SYMBOL(__f_setown);
 
-void f_setown(struct file *filp, unsigned long arg, int force)
+int f_setown(struct file *filp, unsigned long arg, int force)
 {
 	enum pid_type type;
-	struct pid *pid;
-	int who = arg;
+	struct pid *pid = NULL;
+	int who = arg, ret = 0;
+
 	type = PIDTYPE_PID;
 	if (who < 0) {
+		/* avoid overflow below */
+		if (who == INT_MIN)
+			return -EINVAL;
+
 		type = PIDTYPE_PGID;
 		who = -who;
 	}
+
 	rcu_read_lock();
-	pid = find_vpid(who);
-	__f_setown(filp, pid, type, force);
+	if (who) {
+		pid = find_vpid(who);
+		if (!pid)
+			ret = -ESRCH;
+	}
+
+	if (!ret)
+		__f_setown(filp, pid, type, force);
 	rcu_read_unlock();
+
+	return ret;
 }
 EXPORT_SYMBOL(f_setown);
 
@@ -307,6 +321,8 @@ static long fcntl_rw_hint(struct file *file, unsigned int cmd,
 static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		struct file *filp)
 {
+	void __user *argp = (void __user *)arg;
+	struct flock flock;
 	long err = -EINVAL;
 
 	switch (cmd) {
@@ -334,7 +350,11 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 	case F_OFD_GETLK:
 #endif
 	case F_GETLK:
-		err = fcntl_getlk(filp, cmd, (struct flock __user *) arg);
+		if (copy_from_user(&flock, argp, sizeof(flock)))
+			return -EFAULT;
+		err = fcntl_getlk(filp, cmd, &flock);
+		if (!err && copy_to_user(argp, &flock, sizeof(flock)))
+			return -EFAULT;
 		break;
 #if BITS_PER_LONG != 32
 	/* 32-bit arches must use fcntl64() */
@@ -344,7 +364,9 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		/* Fallthrough */
 	case F_SETLK:
 	case F_SETLKW:
-		err = fcntl_setlk(fd, filp, cmd, (struct flock __user *) arg);
+		if (copy_from_user(&flock, argp, sizeof(flock)))
+			return -EFAULT;
+		err = fcntl_setlk(fd, filp, cmd, &flock);
 		break;
 	case F_GETOWN:
 		/*
@@ -358,8 +380,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 		force_successful_syscall_return();
 		break;
 	case F_SETOWN:
-		f_setown(filp, arg, 1);
-		err = 0;
+		err = f_setown(filp, arg, 1);
 		break;
 	case F_GETOWN_EX:
 		err = f_getown_ex(filp, arg);
@@ -450,7 +471,9 @@ out:
 SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd,
 		unsigned long, arg)
 {	
+	void __user *argp = (void __user *)arg;
 	struct fd f = fdget_raw(fd);
+	struct flock64 flock;
 	long err = -EBADF;
 
 	if (!f.file)
@@ -468,14 +491,21 @@ SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd,
 	switch (cmd) {
 	case F_GETLK64:
 	case F_OFD_GETLK:
-		err = fcntl_getlk64(f.file, cmd, (struct flock64 __user *) arg);
+		err = -EFAULT;
+		if (copy_from_user(&flock, argp, sizeof(flock)))
+			break;
+		err = fcntl_getlk64(f.file, cmd, &flock);
+		if (!err && copy_to_user(argp, &flock, sizeof(flock)))
+			err = -EFAULT;
 		break;
 	case F_SETLK64:
 	case F_SETLKW64:
 	case F_OFD_SETLK:
 	case F_OFD_SETLKW:
-		err = fcntl_setlk64(fd, f.file, cmd,
-				(struct flock64 __user *) arg);
+		err = -EFAULT;
+		if (copy_from_user(&flock, argp, sizeof(flock)))
+			break;
+		err = fcntl_setlk64(fd, f.file, cmd, &flock);
 		break;
 	default:
 		err = do_fcntl(fd, cmd, arg, f.file);
@@ -489,57 +519,56 @@ out:
 #endif
 
 #ifdef CONFIG_COMPAT
+/* careful - don't use anywhere else */
+#define copy_flock_fields(from, to)		\
+	(to).l_type = (from).l_type;		\
+	(to).l_whence = (from).l_whence;	\
+	(to).l_start = (from).l_start;		\
+	(to).l_len = (from).l_len;		\
+	(to).l_pid = (from).l_pid;
+
 static int get_compat_flock(struct flock *kfl, struct compat_flock __user *ufl)
 {
-	if (!access_ok(VERIFY_READ, ufl, sizeof(*ufl)) ||
-	    __get_user(kfl->l_type, &ufl->l_type) ||
-	    __get_user(kfl->l_whence, &ufl->l_whence) ||
-	    __get_user(kfl->l_start, &ufl->l_start) ||
-	    __get_user(kfl->l_len, &ufl->l_len) ||
-	    __get_user(kfl->l_pid, &ufl->l_pid))
+	struct compat_flock fl;
+
+	if (copy_from_user(&fl, ufl, sizeof(struct compat_flock)))
 		return -EFAULT;
+	copy_flock_fields(*kfl, fl);
 	return 0;
 }
 
-static int put_compat_flock(struct flock *kfl, struct compat_flock __user *ufl)
+static int get_compat_flock64(struct flock *kfl, struct compat_flock64 __user *ufl)
 {
-	if (!access_ok(VERIFY_WRITE, ufl, sizeof(*ufl)) ||
-	    __put_user(kfl->l_type, &ufl->l_type) ||
-	    __put_user(kfl->l_whence, &ufl->l_whence) ||
-	    __put_user(kfl->l_start, &ufl->l_start) ||
-	    __put_user(kfl->l_len, &ufl->l_len) ||
-	    __put_user(kfl->l_pid, &ufl->l_pid))
+	struct compat_flock64 fl;
+
+	if (copy_from_user(&fl, ufl, sizeof(struct compat_flock64)))
 		return -EFAULT;
+	copy_flock_fields(*kfl, fl);
 	return 0;
 }
 
-#ifndef HAVE_ARCH_GET_COMPAT_FLOCK64
-static int get_compat_flock64(struct flock *kfl, struct compat_flock64 __user *ufl)
+static int put_compat_flock(struct flock *kfl, struct compat_flock __user *ufl)
 {
-	if (!access_ok(VERIFY_READ, ufl, sizeof(*ufl)) ||
-	    __get_user(kfl->l_type, &ufl->l_type) ||
-	    __get_user(kfl->l_whence, &ufl->l_whence) ||
-	    __get_user(kfl->l_start, &ufl->l_start) ||
-	    __get_user(kfl->l_len, &ufl->l_len) ||
-	    __get_user(kfl->l_pid, &ufl->l_pid))
+	struct compat_flock fl;
+
+	memset(&fl, 0, sizeof(struct compat_flock));
+	copy_flock_fields(fl, *kfl);
+	if (copy_to_user(ufl, &fl, sizeof(struct compat_flock)))
 		return -EFAULT;
 	return 0;
 }
-#endif
 
-#ifndef HAVE_ARCH_PUT_COMPAT_FLOCK64
 static int put_compat_flock64(struct flock *kfl, struct compat_flock64 __user *ufl)
 {
-	if (!access_ok(VERIFY_WRITE, ufl, sizeof(*ufl)) ||
-	    __put_user(kfl->l_type, &ufl->l_type) ||
-	    __put_user(kfl->l_whence, &ufl->l_whence) ||
-	    __put_user(kfl->l_start, &ufl->l_start) ||
-	    __put_user(kfl->l_len, &ufl->l_len) ||
-	    __put_user(kfl->l_pid, &ufl->l_pid))
+	struct compat_flock64 fl;
+
+	memset(&fl, 0, sizeof(struct compat_flock64));
+	copy_flock_fields(fl, *kfl);
+	if (copy_to_user(ufl, &fl, sizeof(struct compat_flock64)))
 		return -EFAULT;
 	return 0;
 }
-#endif
+#undef copy_flock_fields
 
 static unsigned int
 convert_fcntl_cmd(unsigned int cmd)
@@ -556,76 +585,92 @@ convert_fcntl_cmd(unsigned int cmd)
 	return cmd;
 }
 
+/*
+ * GETLK was successful and we need to return the data, but it needs to fit in
+ * the compat structure.
+ * l_start shouldn't be too big, unless the original start + end is greater than
+ * COMPAT_OFF_T_MAX, in which case the app was asking for trouble, so we return
+ * -EOVERFLOW in that case.  l_len could be too big, in which case we just
+ * truncate it, and only allow the app to see that part of the conflicting lock
+ * that might make sense to it anyway
+ */
+static int fixup_compat_flock(struct flock *flock)
+{
+	if (flock->l_start > COMPAT_OFF_T_MAX)
+		return -EOVERFLOW;
+	if (flock->l_len > COMPAT_OFF_T_MAX)
+		flock->l_len = COMPAT_OFF_T_MAX;
+	return 0;
+}
+
 COMPAT_SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd,
 		       compat_ulong_t, arg)
 {
-	mm_segment_t old_fs;
-	struct flock f;
-	long ret;
-	unsigned int conv_cmd;
+	struct fd f = fdget_raw(fd);
+	struct flock flock;
+	long err = -EBADF;
+
+	if (!f.file)
+		return err;
+
+	if (unlikely(f.file->f_mode & FMODE_PATH)) {
+		if (!check_fcntl_cmd(cmd))
+			goto out_put;
+	}
+
+	err = security_file_fcntl(f.file, cmd, arg);
+	if (err)
+		goto out_put;
 
 	switch (cmd) {
 	case F_GETLK:
+		err = get_compat_flock(&flock, compat_ptr(arg));
+		if (err)
+			break;
+		err = fcntl_getlk(f.file, convert_fcntl_cmd(cmd), &flock);
+		if (err)
+			break;
+		err = fixup_compat_flock(&flock);
+		if (err)
+			return err;
+		err = put_compat_flock(&flock, compat_ptr(arg));
+		break;
+	case F_GETLK64:
+	case F_OFD_GETLK:
+		err = get_compat_flock64(&flock, compat_ptr(arg));
+		if (err)
+			break;
+		err = fcntl_getlk(f.file, convert_fcntl_cmd(cmd), &flock);
+		if (err)
+			break;
+		err = fixup_compat_flock(&flock);
+		if (err)
+			return err;
+		err = put_compat_flock64(&flock, compat_ptr(arg));
+		break;
 	case F_SETLK:
 	case F_SETLKW:
-		ret = get_compat_flock(&f, compat_ptr(arg));
-		if (ret != 0)
+		err = get_compat_flock(&flock, compat_ptr(arg));
+		if (err)
 			break;
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		ret = sys_fcntl(fd, cmd, (unsigned long)&f);
-		set_fs(old_fs);
-		if (cmd == F_GETLK && ret == 0) {
-			/* GETLK was successful and we need to return the data...
-			 * but it needs to fit in the compat structure.
-			 * l_start shouldn't be too big, unless the original
-			 * start + end is greater than COMPAT_OFF_T_MAX, in which
-			 * case the app was asking for trouble, so we return
-			 * -EOVERFLOW in that case.
-			 * l_len could be too big, in which case we just truncate it,
-			 * and only allow the app to see that part of the conflicting
-			 * lock that might make sense to it anyway
-			 */
-
-			if (f.l_start > COMPAT_OFF_T_MAX)
-				ret = -EOVERFLOW;
-			if (f.l_len > COMPAT_OFF_T_MAX)
-				f.l_len = COMPAT_OFF_T_MAX;
-			if (ret == 0)
-				ret = put_compat_flock(&f, compat_ptr(arg));
-		}
+		err = fcntl_setlk(fd, f.file, convert_fcntl_cmd(cmd), &flock);
 		break;
-
-	case F_GETLK64:
 	case F_SETLK64:
 	case F_SETLKW64:
-	case F_OFD_GETLK:
 	case F_OFD_SETLK:
 	case F_OFD_SETLKW:
-		ret = get_compat_flock64(&f, compat_ptr(arg));
-		if (ret != 0)
+		err = get_compat_flock64(&flock, compat_ptr(arg));
+		if (err)
 			break;
-		old_fs = get_fs();
-		set_fs(KERNEL_DS);
-		conv_cmd = convert_fcntl_cmd(cmd);
-		ret = sys_fcntl(fd, conv_cmd, (unsigned long)&f);
-		set_fs(old_fs);
-		if ((conv_cmd == F_GETLK || conv_cmd == F_OFD_GETLK) && ret == 0) {
-			/* need to return lock information - see above for commentary */
-			if (f.l_start > COMPAT_LOFF_T_MAX)
-				ret = -EOVERFLOW;
-			if (f.l_len > COMPAT_LOFF_T_MAX)
-				f.l_len = COMPAT_LOFF_T_MAX;
-			if (ret == 0)
-				ret = put_compat_flock64(&f, compat_ptr(arg));
-		}
+		err = fcntl_setlk(fd, f.file, convert_fcntl_cmd(cmd), &flock);
 		break;
-
 	default:
-		ret = sys_fcntl(fd, cmd, arg);
+		err = do_fcntl(fd, cmd, arg, f.file);
 		break;
 	}
-	return ret;
+out_put:
+	fdput(f);
+	return err;
 }
 
 COMPAT_SYSCALL_DEFINE3(fcntl, unsigned int, fd, unsigned int, cmd,

fs/locks.c

@@ -1858,8 +1858,8 @@ EXPORT_SYMBOL(generic_setlease);
  *
  * Call this to establish a lease on the file. The "lease" argument is not
  * used for F_UNLCK requests and may be NULL. For commands that set or alter
- * an existing lease, the (*lease)->fl_lmops->lm_break operation must be set;
- * if not, this function will return -ENOLCK (and generate a scary-looking
+ * an existing lease, the ``(*lease)->fl_lmops->lm_break`` operation must be
+ * set; if not, this function will return -ENOLCK (and generate a scary-looking
  * stack trace).
  *
  * The "priv" pointer is passed directly to the lm_setup function as-is. It
@@ -1972,15 +1972,13 @@ EXPORT_SYMBOL(locks_lock_inode_wait);
  *	@cmd: the type of lock to apply.
  *
  *	Apply a %FL_FLOCK style lock to an open file descriptor.
- *	The @cmd can be one of
+ *	The @cmd can be one of:
  *
- *	%LOCK_SH -- a shared lock.
- *
- *	%LOCK_EX -- an exclusive lock.
- *
- *	%LOCK_UN -- remove an existing lock.
- *
- *	%LOCK_MAND -- a `mandatory' flock.  This exists to emulate Windows Share Modes.
+ *	- %LOCK_SH -- a shared lock.
+ *	- %LOCK_EX -- an exclusive lock.
+ *	- %LOCK_UN -- remove an existing lock.
+ *	- %LOCK_MAND -- a 'mandatory' flock.
+ *	  This exists to emulate Windows Share Modes.
  *
  *	%LOCK_MAND can be combined with %LOCK_READ or %LOCK_WRITE to allow other
  *	processes read and write access respectively.
@@ -2086,26 +2084,22 @@ static void posix_lock_to_flock64(struct flock64 *flock, struct file_lock *fl)
 /* Report the first existing lock that would conflict with l.
  * This implements the F_GETLK command of fcntl().
  */
-int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock __user *l)
+int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock *flock)
 {
 	struct file_lock file_lock;
-	struct flock flock;
 	int error;
 
-	error = -EFAULT;
-	if (copy_from_user(&flock, l, sizeof(flock)))
-		goto out;
 	error = -EINVAL;
-	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
+	if (flock->l_type != F_RDLCK && flock->l_type != F_WRLCK)
 		goto out;
 
-	error = flock_to_posix_lock(filp, &file_lock, &flock);
+	error = flock_to_posix_lock(filp, &file_lock, flock);
 	if (error)
 		goto out;
 
 	if (cmd == F_OFD_GETLK) {
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_GETLK;
@@ -2117,15 +2111,12 @@ int fcntl_getlk(struct file *filp, unsigned int cmd, struct flock __user *l)
 	if (error)
 		goto out;
  
-	flock.l_type = file_lock.fl_type;
+	flock->l_type = file_lock.fl_type;
 	if (file_lock.fl_type != F_UNLCK) {
-		error = posix_lock_to_flock(&flock, &file_lock);
+		error = posix_lock_to_flock(flock, &file_lock);
 		if (error)
 			goto rel_priv;
 	}
-	error = -EFAULT;
-	if (!copy_to_user(l, &flock, sizeof(flock)))
-		error = 0;
 rel_priv:
 	locks_release_private(&file_lock);
 out:
@@ -2218,26 +2209,16 @@ check_fmode_for_setlk(struct file_lock *fl)
  * This implements both the F_SETLK and F_SETLKW commands of fcntl().
  */
 int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
-		struct flock __user *l)
+		struct flock *flock)
 {
 	struct file_lock *file_lock = locks_alloc_lock();
-	struct flock flock;
-	struct inode *inode;
+	struct inode *inode = locks_inode(filp);
 	struct file *f;
 	int error;
 
 	if (file_lock == NULL)
 		return -ENOLCK;
 
-	inode = locks_inode(filp);
-
-	/*
-	 * This might block, so we do it before checking the inode.
-	 */
-	error = -EFAULT;
-	if (copy_from_user(&flock, l, sizeof(flock)))
-		goto out;
-
 	/* Don't allow mandatory locks on files that may be memory mapped
 	 * and shared.
 	 */
@@ -2246,7 +2227,7 @@ int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
 		goto out;
 	}
 
-	error = flock_to_posix_lock(filp, file_lock, &flock);
+	error = flock_to_posix_lock(filp, file_lock, flock);
 	if (error)
 		goto out;
 
@@ -2261,7 +2242,7 @@ int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
 	switch (cmd) {
 	case F_OFD_SETLK:
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_SETLK;
@@ -2270,7 +2251,7 @@ int fcntl_setlk(unsigned int fd, struct file *filp, unsigned int cmd,
 		break;
 	case F_OFD_SETLKW:
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_SETLKW;
@@ -2315,26 +2296,22 @@ out:
 /* Report the first existing lock that would conflict with l.
  * This implements the F_GETLK command of fcntl().
  */
-int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l)
+int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 *flock)
 {
 	struct file_lock file_lock;
-	struct flock64 flock;
 	int error;
 
-	error = -EFAULT;
-	if (copy_from_user(&flock, l, sizeof(flock)))
-		goto out;
 	error = -EINVAL;
-	if ((flock.l_type != F_RDLCK) && (flock.l_type != F_WRLCK))
+	if (flock->l_type != F_RDLCK && flock->l_type != F_WRLCK)
 		goto out;
 
-	error = flock64_to_posix_lock(filp, &file_lock, &flock);
+	error = flock64_to_posix_lock(filp, &file_lock, flock);
 	if (error)
 		goto out;
 
 	if (cmd == F_OFD_GETLK) {
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_GETLK64;
@@ -2346,13 +2323,9 @@ int fcntl_getlk64(struct file *filp, unsigned int cmd, struct flock64 __user *l)
 	if (error)
 		goto out;
 
-	flock.l_type = file_lock.fl_type;
+	flock->l_type = file_lock.fl_type;
 	if (file_lock.fl_type != F_UNLCK)
-		posix_lock_to_flock64(&flock, &file_lock);
-
-	error = -EFAULT;
-	if (!copy_to_user(l, &flock, sizeof(flock)))
-		error = 0;
+		posix_lock_to_flock64(flock, &file_lock);
 
 	locks_release_private(&file_lock);
 out:
@@ -2363,26 +2336,16 @@ out:
  * This implements both the F_SETLK and F_SETLKW commands of fcntl().
  */
 int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
-		struct flock64 __user *l)
+		struct flock64 *flock)
 {
 	struct file_lock *file_lock = locks_alloc_lock();
-	struct flock64 flock;
-	struct inode *inode;
+	struct inode *inode = locks_inode(filp);
 	struct file *f;
 	int error;
 
 	if (file_lock == NULL)
 		return -ENOLCK;
 
-	/*
-	 * This might block, so we do it before checking the inode.
-	 */
-	error = -EFAULT;
-	if (copy_from_user(&flock, l, sizeof(flock)))
-		goto out;
-
-	inode = locks_inode(filp);
-
 	/* Don't allow mandatory locks on files that may be memory mapped
 	 * and shared.
 	 */
@@ -2391,7 +2354,7 @@ int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
 		goto out;
 	}
 
-	error = flock64_to_posix_lock(filp, file_lock, &flock);
+	error = flock64_to_posix_lock(filp, file_lock, flock);
 	if (error)
 		goto out;
 
@@ -2406,7 +2369,7 @@ int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
 	switch (cmd) {
 	case F_OFD_SETLK:
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_SETLK64;
@@ -2415,7 +2378,7 @@ int fcntl_setlk64(unsigned int fd, struct file *filp, unsigned int cmd,
 		break;
 	case F_OFD_SETLKW:
 		error = -EINVAL;
-		if (flock.l_pid != 0)
+		if (flock->l_pid != 0)
 			goto out;
 
 		cmd = F_SETLKW64;

fs/nfsd/vfs.c

@@ -1464,41 +1464,34 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp,
 __be32
 nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp)
 {
-	mm_segment_t	oldfs;
 	__be32		err;
-	int		host_err;
+	const char *link;
 	struct path path;
+	DEFINE_DELAYED_CALL(done);
+	int len;
 
 	err = fh_verify(rqstp, fhp, S_IFLNK, NFSD_MAY_NOP);
-	if (err)
-		goto out;
+	if (unlikely(err))
+		return err;
 
 	path.mnt = fhp->fh_export->ex_path.mnt;
 	path.dentry = fhp->fh_dentry;
 
-	err = nfserr_inval;
-	if (!d_is_symlink(path.dentry))
-		goto out;
+	if (unlikely(!d_is_symlink(path.dentry)))
+		return nfserr_inval;
 
 	touch_atime(&path);
-	/* N.B. Why does this call need a get_fs()??
-	 * Remove the set_fs and watch the fireworks:-) --okir
-	 */
 
-	oldfs = get_fs(); set_fs(KERNEL_DS);
-	host_err = vfs_readlink(path.dentry, (char __user *)buf, *lenp);
-	set_fs(oldfs);
+	link = vfs_get_link(path.dentry, &done);
+	if (IS_ERR(link))
+		return nfserrno(PTR_ERR(link));
 
-	if (host_err < 0)
-		goto out_nfserr;
-	*lenp = host_err;
-	err = 0;
-out:
-	return err;
-
-out_nfserr:
-	err = nfserrno(host_err);
-	goto out;
+	len = strlen(link);
+	if (len < *lenp)
+		*lenp = len;
+	memcpy(buf, link, *lenp);
+	do_delayed_call(&done);
+	return 0;
 }
 
 /*

fs/statfs.c

@@ -244,6 +244,7 @@ SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf)
 #ifdef CONFIG_COMPAT
 static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs *kbuf)
 {
+	struct compat_statfs buf;
 	if (sizeof ubuf->f_blocks == 4) {
 		if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail |
 		     kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL)
@@ -257,20 +258,20 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs *
 		 && (kbuf->f_ffree & 0xffffffff00000000ULL))
 			return -EOVERFLOW;
 	}
-	if (!access_ok(VERIFY_WRITE, ubuf, sizeof(*ubuf)) ||
-	    __put_user(kbuf->f_type, &ubuf->f_type) ||
-	    __put_user(kbuf->f_bsize, &ubuf->f_bsize) ||
-	    __put_user(kbuf->f_blocks, &ubuf->f_blocks) ||
-	    __put_user(kbuf->f_bfree, &ubuf->f_bfree) ||
-	    __put_user(kbuf->f_bavail, &ubuf->f_bavail) ||
-	    __put_user(kbuf->f_files, &ubuf->f_files) ||
-	    __put_user(kbuf->f_ffree, &ubuf->f_ffree) ||
-	    __put_user(kbuf->f_namelen, &ubuf->f_namelen) ||
-	    __put_user(kbuf->f_fsid.val[0], &ubuf->f_fsid.val[0]) ||
-	    __put_user(kbuf->f_fsid.val[1], &ubuf->f_fsid.val[1]) ||
-	    __put_user(kbuf->f_frsize, &ubuf->f_frsize) ||
-	    __put_user(kbuf->f_flags, &ubuf->f_flags) ||
-	    __clear_user(ubuf->f_spare, sizeof(ubuf->f_spare)))
+	memset(&buf, 0, sizeof(struct compat_statfs));
+	buf.f_type = kbuf->f_type;
+	buf.f_bsize = kbuf->f_bsize;
+	buf.f_blocks = kbuf->f_blocks;
+	buf.f_bfree = kbuf->f_bfree;
+	buf.f_bavail = kbuf->f_bavail;
+	buf.f_files = kbuf->f_files;
+	buf.f_ffree = kbuf->f_ffree;
+	buf.f_namelen = kbuf->f_namelen;
+	buf.f_fsid.val[0] = kbuf->f_fsid.val[0];
+	buf.f_fsid.val[1] = kbuf->f_fsid.val[1];
+	buf.f_frsize = kbuf->f_frsize;
+	buf.f_flags = kbuf->f_flags;
+	if (copy_to_user(ubuf, &buf, sizeof(struct compat_statfs)))
 		return -EFAULT;
 	return 0;
 }
@@ -299,6 +300,7 @@ COMPAT_SYSCALL_DEFINE2(fstatfs, unsigned int, fd, struct compat_statfs __user *,
 
 static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstatfs *kbuf)
 {
+	struct compat_statfs64 buf;
 	if (sizeof(ubuf->f_bsize) == 4) {
 		if ((kbuf->f_type | kbuf->f_bsize | kbuf->f_namelen |
 		     kbuf->f_frsize | kbuf->f_flags) & 0xffffffff00000000ULL)
@@ -312,20 +314,20 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat
 		 && (kbuf->f_ffree & 0xffffffff00000000ULL))
 			return -EOVERFLOW;
 	}
-	if (!access_ok(VERIFY_WRITE, ubuf, sizeof(*ubuf)) ||
-	    __put_user(kbuf->f_type, &ubuf->f_type) ||
-	    __put_user(kbuf->f_bsize, &ubuf->f_bsize) ||
-	    __put_user(kbuf->f_blocks, &ubuf->f_blocks) ||
-	    __put_user(kbuf->f_bfree, &ubuf->f_bfree) ||
-	    __put_user(kbuf->f_bavail, &ubuf->f_bavail) ||
-	    __put_user(kbuf->f_files, &ubuf->f_files) ||
-	    __put_user(kbuf->f_ffree, &ubuf->f_ffree) ||
-	    __put_user(kbuf->f_namelen, &ubuf->f_namelen) ||
-	    __put_user(kbuf->f_fsid.val[0], &ubuf->f_fsid.val[0]) ||
-	    __put_user(kbuf->f_fsid.val[1], &ubuf->f_fsid.val[1]) ||
-	    __put_user(kbuf->f_frsize, &ubuf->f_frsize) ||
-	    __put_user(kbuf->f_flags, &ubuf->f_flags) ||
-	    __clear_user(ubuf->f_spare, sizeof(ubuf->f_spare)))
+	memset(&buf, 0, sizeof(struct compat_statfs64));
+	buf.f_type = kbuf->f_type;
+	buf.f_bsize = kbuf->f_bsize;
+	buf.f_blocks = kbuf->f_blocks;
+	buf.f_bfree = kbuf->f_bfree;
+	buf.f_bavail = kbuf->f_bavail;
+	buf.f_files = kbuf->f_files;
+	buf.f_ffree = kbuf->f_ffree;
+	buf.f_namelen = kbuf->f_namelen;
+	buf.f_fsid.val[0] = kbuf->f_fsid.val[0];
+	buf.f_fsid.val[1] = kbuf->f_fsid.val[1];
+	buf.f_frsize = kbuf->f_frsize;
+	buf.f_flags = kbuf->f_flags;
+	if (copy_to_user(ubuf, &buf, sizeof(struct compat_statfs64)))
 		return -EFAULT;
 	return 0;
 }

include/linux/fs.h

@@ -1047,14 +1047,14 @@ static inline struct inode *locks_inode(const struct file *f)
 }
 
 #ifdef CONFIG_FILE_LOCKING
-extern int fcntl_getlk(struct file *, unsigned int, struct flock __user *);
+extern int fcntl_getlk(struct file *, unsigned int, struct flock *);
 extern int fcntl_setlk(unsigned int, struct file *, unsigned int,
-			struct flock __user *);
+			struct flock *);
 
 #if BITS_PER_LONG == 32
-extern int fcntl_getlk64(struct file *, unsigned int, struct flock64 __user *);
+extern int fcntl_getlk64(struct file *, unsigned int, struct flock64 *);
 extern int fcntl_setlk64(unsigned int, struct file *, unsigned int,
-			struct flock64 __user *);
+			struct flock64 *);
 #endif
 
 extern int fcntl_setlease(unsigned int fd, struct file *filp, long arg);
@@ -1258,7 +1258,7 @@ extern void fasync_free(struct fasync_struct *);
 extern void kill_fasync(struct fasync_struct **, int, int);
 
 extern void __f_setown(struct file *filp, struct pid *, enum pid_type, int force);
-extern void f_setown(struct file *filp, unsigned long arg, int force);
+extern int f_setown(struct file *filp, unsigned long arg, int force);
 extern void f_delown(struct file *filp);
 extern pid_t f_getown(struct file *filp);
 extern int send_sigurg(struct fown_struct *fown);

net/socket.c

@@ -991,8 +991,7 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 			err = -EFAULT;
 			if (get_user(pid, (int __user *)argp))
 				break;
-			f_setown(sock->file, pid, 1);
-			err = 0;
+			err = f_setown(sock->file, pid, 1);
 			break;
 		case FIOGETOWN:
 		case SIOCGPGRP: