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

Pull second vfs pile from Al Viro:
"The stuff in there: fsfreeze deadlock fixes by Jan (essentially, the
deadlock reproduced by xfstests 068), symlink and hardlink restriction
patches, plus assorted cleanups and fixes.

Note that another fsfreeze deadlock (emergency thaw one) is *not*
dealt with - the series by Fernando conflicts a lot with Jan's, breaks
userland ABI (FIFREEZE semantics gets changed) and trades the deadlock
for massive vfsmount leak; this is going to be handled next cycle.
There probably will be another pull request, but that stuff won't be
in it."

Fix up trivial conflicts due to unrelated changes next to each other in
drivers/{staging/gdm72xx/usb_boot.c, usb/gadget/storage_common.c}

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (54 commits)
delousing target_core_file a bit
Documentation: Correct s_umount state for freeze_fs/unfreeze_fs
fs: Remove old freezing mechanism
ext2: Implement freezing
btrfs: Convert to new freezing mechanism
nilfs2: Convert to new freezing mechanism
ntfs: Convert to new freezing mechanism
fuse: Convert to new freezing mechanism
gfs2: Convert to new freezing mechanism
ocfs2: Convert to new freezing mechanism
xfs: Convert to new freezing code
ext4: Convert to new freezing mechanism
fs: Protect write paths by sb_start_write - sb_end_write
fs: Skip atime update on frozen filesystem
fs: Add freezing handling to mnt_want_write() / mnt_drop_write()
fs: Improve filesystem freezing handling
switch the protection of percpu_counter list to spinlock
nfsd: Push mnt_want_write() outside of i_mutex
btrfs: Push mnt_want_write() outside of i_mutex
fat: Push mnt_want_write() outside of i_mutex
...

Documentation/filesystems/Locking

···
 put_super:		write
 write_super:		read
 sync_fs:		read
-freeze_fs:		read
-unfreeze_fs:		read
+freeze_fs:		write
+unfreeze_fs:		write
 statfs:			maybe(read)	(see below)
 remount_fs:		write
 umount_begin:		no

Documentation/sysctl/fs.txt

···
 - nr_open
 - overflowuid
 - overflowgid
+- protected_hardlinks
+- protected_symlinks
 - suid_dumpable
 - super-max
 - super-nr
···
 
 These sysctls allow you to change the value of the fixed UID and GID.
 The default is 65534.
+
+==============================================================
+
+protected_hardlinks:
+
+A long-standing class of security issues is the hardlink-based
+time-of-check-time-of-use race, most commonly seen in world-writable
+directories like /tmp. The common method of exploitation of this flaw
+is to cross privilege boundaries when following a given hardlink (i.e. a
+root process follows a hardlink created by another user). Additionally,
+on systems without separated partitions, this stops unauthorized users
+from "pinning" vulnerable setuid/setgid files against being upgraded by
+the administrator, or linking to special files.
+
+When set to "0", hardlink creation behavior is unrestricted.
+
+When set to "1" hardlinks cannot be created by users if they do not
+already own the source file, or do not have read/write access to it.
+
+This protection is based on the restrictions in Openwall and grsecurity.
+
+==============================================================
+
+protected_symlinks:
+
+A long-standing class of security issues is the symlink-based
+time-of-check-time-of-use race, most commonly seen in world-writable
+directories like /tmp. The common method of exploitation of this flaw
+is to cross privilege boundaries when following a given symlink (i.e. a
+root process follows a symlink belonging to another user). For a likely
+incomplete list of hundreds of examples across the years, please see:
+http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp
+
+When set to "0", symlink following behavior is unrestricted.
+
+When set to "1" symlinks are permitted to be followed only when outside
+a sticky world-writable directory, or when the uid of the symlink and
+follower match, or when the directory owner matches the symlink's owner.
+
+This protection is based on the restrictions in Openwall and grsecurity.
 
 ==============================================================
 

arch/powerpc/platforms/cell/spufs/inode.c

···
 static int spufs_rmdir(struct inode *parent, struct dentry *dir)
 {
 	/* remove all entries */
+	int res;
 	spufs_prune_dir(dir);
 	d_drop(dir);
-
-	return simple_rmdir(parent, dir);
+	res = simple_rmdir(parent, dir);
+	/* We have to give up the mm_struct */
+	spu_forget(SPUFS_I(dir->d_inode)->i_ctx);
+	return res;
 }
 
 static int spufs_fill_dir(struct dentry *dir,
···
 	ret = spufs_rmdir(parent, dir);
 	mutex_unlock(&parent->i_mutex);
 	WARN_ON(ret);
-
-	/* We have to give up the mm_struct */
-	spu_forget(ctx);
 
 	return dcache_dir_close(inode, file);
 }
···
 	struct spu_context *neighbor;
 	struct path path = {.mnt = mnt, .dentry = dentry};
 
-	ret = -EPERM;
 	if ((flags & SPU_CREATE_NOSCHED) &&
 	    !capable(CAP_SYS_NICE))
-		goto out_unlock;
+		return -EPERM;
 
-	ret = -EINVAL;
 	if ((flags & (SPU_CREATE_NOSCHED | SPU_CREATE_ISOLATE))
 	    == SPU_CREATE_ISOLATE)
-		goto out_unlock;
+		return -EINVAL;
 
-	ret = -ENODEV;
 	if ((flags & SPU_CREATE_ISOLATE) && !isolated_loader)
-		goto out_unlock;
+		return -ENODEV;
 
 	gang = NULL;
 	neighbor = NULL;
 	affinity = flags & (SPU_CREATE_AFFINITY_MEM | SPU_CREATE_AFFINITY_SPU);
 	if (affinity) {
 		gang = SPUFS_I(inode)->i_gang;
-		ret = -EINVAL;
 		if (!gang)
-			goto out_unlock;
+			return -EINVAL;
 		mutex_lock(&gang->aff_mutex);
 		neighbor = spufs_assert_affinity(flags, gang, aff_filp);
 		if (IS_ERR(neighbor)) {
···
 	}
 
 	ret = spufs_context_open(&path);
-	if (ret < 0) {
+	if (ret < 0)
 		WARN_ON(spufs_rmdir(inode, dentry));
-		if (affinity)
-			mutex_unlock(&gang->aff_mutex);
-		mutex_unlock(&inode->i_mutex);
-		spu_forget(SPUFS_I(dentry->d_inode)->i_ctx);
-		goto out;
-	}
 
 out_aff_unlock:
 	if (affinity)
 		mutex_unlock(&gang->aff_mutex);
-out_unlock:
-	mutex_unlock(&inode->i_mutex);
-out:
-	dput(dentry);
 	return ret;
 }
 
···
 	int ret;
 
 	ret = spufs_mkgang(inode, dentry, mode & S_IRWXUGO);
-	if (ret)
-		goto out;
-
-	ret = spufs_gang_open(&path);
-	if (ret < 0) {
-		int err = simple_rmdir(inode, dentry);
-		WARN_ON(err);
+	if (!ret) {
+		ret = spufs_gang_open(&path);
+		if (ret < 0) {
+			int err = simple_rmdir(inode, dentry);
+			WARN_ON(err);
+		}
 	}
-
-out:
-	mutex_unlock(&inode->i_mutex);
-	dput(dentry);
 	return ret;
 }
 
···
 long spufs_create(struct path *path, struct dentry *dentry,
 		unsigned int flags, umode_t mode, struct file *filp)
 {
+	struct inode *dir = path->dentry->d_inode;
 	int ret;
 
-	ret = -EINVAL;
 	/* check if we are on spufs */
 	if (path->dentry->d_sb->s_type != &spufs_type)
-		goto out;
+		return -EINVAL;
 
 	/* don't accept undefined flags */
 	if (flags & (~SPU_CREATE_FLAG_ALL))
-		goto out;
+		return -EINVAL;
 
 	/* only threads can be underneath a gang */
-	if (path->dentry != path->dentry->d_sb->s_root) {
-		if ((flags & SPU_CREATE_GANG) ||
-		    !SPUFS_I(path->dentry->d_inode)->i_gang)
-			goto out;
-	}
+	if (path->dentry != path->dentry->d_sb->s_root)
+		if ((flags & SPU_CREATE_GANG) || !SPUFS_I(dir)->i_gang)
+			return -EINVAL;
 
 	mode &= ~current_umask();
 
 	if (flags & SPU_CREATE_GANG)
-		ret = spufs_create_gang(path->dentry->d_inode,
-					 dentry, path->mnt, mode);
+		ret = spufs_create_gang(dir, dentry, path->mnt, mode);
 	else
-		ret = spufs_create_context(path->dentry->d_inode,
-					    dentry, path->mnt, flags, mode,
+		ret = spufs_create_context(dir, dentry, path->mnt, flags, mode,
 					    filp);
 	if (ret >= 0)
-		fsnotify_mkdir(path->dentry->d_inode, dentry);
-	return ret;
+		fsnotify_mkdir(dir, dentry);
 
-out:
-	mutex_unlock(&path->dentry->d_inode->i_mutex);
-	dput(dentry);
 	return ret;
 }
 

arch/powerpc/platforms/cell/spufs/syscalls.c

···
 	ret = PTR_ERR(dentry);
 	if (!IS_ERR(dentry)) {
 		ret = spufs_create(&path, dentry, flags, mode, neighbor);
-		path_put(&path);
+		done_path_create(&path, dentry);
 	}
 
 	return ret;

drivers/base/devtmpfs.c

···
 	if (!err)
 		/* mark as kernel-created inode */
 		dentry->d_inode->i_private = &thread;
-	dput(dentry);
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
+	done_path_create(&path, dentry);
 	return err;
 }
 
···
 		/* mark as kernel-created inode */
 		dentry->d_inode->i_private = &thread;
 	}
-	dput(dentry);
-
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
+	done_path_create(&path, dentry);
 	return err;
 }
 

drivers/net/wireless/brcm80211/brcmfmac/dhd_linux.c

···
 	kfree(buf);
 	/* close file before return */
 	if (fp)
-		filp_close(fp, current->files);
+		filp_close(fp, NULL);
 	/* restore previous address limit */
 	set_fs(old_fs);
 

drivers/staging/bcm/Misc.c

···
 
 static struct file *open_firmware_file(struct bcm_mini_adapter *Adapter, const char *path)
 {
-	struct file *flp = NULL;
-	mm_segment_t oldfs;
-	oldfs = get_fs();
-	set_fs(get_ds());
-	flp = filp_open(path, O_RDONLY, S_IRWXU);
-	set_fs(oldfs);
+	struct file *flp = filp_open(path, O_RDONLY, S_IRWXU);
 	if (IS_ERR(flp)) {
 		pr_err(DRV_NAME "Unable To Open File %s, err %ld", path, PTR_ERR(flp));
 		flp = NULL;
···
 {
 	int errorno = 0;
 	struct file *flp = NULL;
-	mm_segment_t oldfs;
 	struct timeval tv = {0};
 
 	flp = open_firmware_file(Adapter, path);
 	if (!flp) {
-		errorno = -ENOENT;
 		BCM_DEBUG_PRINT(Adapter, DBG_TYPE_INITEXIT, MP_INIT, DBG_LVL_ALL, "Unable to Open %s\n", path);
-		goto exit_download;
+		return -ENOENT;
 	}
 	BCM_DEBUG_PRINT(Adapter, DBG_TYPE_INITEXIT, MP_INIT, DBG_LVL_ALL, "Opened file is = %s and length =0x%lx to be downloaded at =0x%x", path, (unsigned long)flp->f_dentry->d_inode->i_size, loc);
 	do_gettimeofday(&tv);
···
 		errorno = -EIO;
 		goto exit_download;
 	}
-	oldfs = get_fs();
-	set_fs(get_ds());
 	vfs_llseek(flp, 0, 0);
-	set_fs(oldfs);
 	if (Adapter->bcm_file_readback_from_chip(Adapter->pvInterfaceAdapter, flp, loc)) {
 		BCM_DEBUG_PRINT(Adapter, DBG_TYPE_INITEXIT, MP_INIT, DBG_LVL_ALL, "Failed to read back firmware!");
 		errorno = -EIO;
···
 	}
 
 exit_download:
-	oldfs = get_fs();
-	set_fs(get_ds());
-	if (flp && !(IS_ERR(flp)))
-		filp_close(flp, current->files);
-	set_fs(oldfs);
-
+	filp_close(flp, NULL);
 	return errorno;
 }
 
···
 static int bcm_parse_target_params(struct bcm_mini_adapter *Adapter)
 {
 	struct file *flp = NULL;
-	mm_segment_t oldfs = {0};
 	char *buff;
 	int len = 0;
-	loff_t pos = 0;
 
 	buff = kmalloc(BUFFER_1K, GFP_KERNEL);
 	if (!buff)
···
 		Adapter->pstargetparams = NULL;
 		return -ENOENT;
 	}
-	oldfs = get_fs();
-	set_fs(get_ds());
-	len = vfs_read(flp, (void __user __force *)buff, BUFFER_1K, &pos);
-	set_fs(oldfs);
+	len = kernel_read(flp, 0, buff, BUFFER_1K);
+	filp_close(flp, NULL);
 
 	if (len != sizeof(STARGETPARAMS)) {
 		BCM_DEBUG_PRINT(Adapter, DBG_TYPE_INITEXIT, MP_INIT, DBG_LVL_ALL, "Mismatch in Target Param Structure!\n");
 		kfree(buff);
 		kfree(Adapter->pstargetparams);
 		Adapter->pstargetparams = NULL;
-		filp_close(flp, current->files);
 		return -ENOENT;
 	}
-	filp_close(flp, current->files);
 
 	/* Check for autolink in config params */
 	/*

drivers/staging/gdm72xx/sdio_boot.c

···
 		return -ENOENT;
 	}
 
-	if (filp->f_dentry)
-		inode = filp->f_dentry->d_inode;
-	if (!inode || !S_ISREG(inode->i_mode)) {
+	inode = filp->f_dentry->d_inode;
+	if (!S_ISREG(inode->i_mode)) {
 		printk(KERN_ERR "Invalid file type: %s\n", img_name);
 		ret = -EINVAL;
 		goto out;
···
 		pno++;
 	}
 out:
-	filp_close(filp, current->files);
+	filp_close(filp, NULL);
 	return ret;
 }
 

drivers/staging/gdm72xx/usb_boot.c

···
 	filp = filp_open(img_name, O_RDONLY | O_LARGEFILE, 0);
 	if (IS_ERR(filp)) {
 		printk(KERN_ERR "Can't find %s.\n", img_name);
-		set_fs(fs);
 		ret = PTR_ERR(filp);
 		goto restore_fs;
 	}
 
-	if (filp->f_dentry)
-		inode = filp->f_dentry->d_inode;
-	if (!inode || !S_ISREG(inode->i_mode)) {
+	inode = filp->f_dentry->d_inode;
+	if (!S_ISREG(inode->i_mode)) {
 		printk(KERN_ERR "Invalid file type: %s\n", img_name);
 		ret = -EINVAL;
 		goto out;
···
 		ret = -EINVAL;
 	}
 out:
-	filp_close(filp, current->files);
+	filp_close(filp, NULL);
 
 restore_fs:
 	set_fs(fs);
···
 		goto restore_fs;
 	}
 
-	if (filp->f_dentry) {
-		inode = filp->f_dentry->d_inode;
-		if (!inode || !S_ISREG(inode->i_mode)) {
-			printk(KERN_ERR "Invalid file type: %s\n", path);
-			ret = -EINVAL;
-			goto out;
-		}
+	inode = filp->f_dentry->d_inode;
+	if (!S_ISREG(inode->i_mode)) {
+		printk(KERN_ERR "Invalid file type: %s\n", path);
+		ret = -EINVAL;
+		goto out;
 	}
 
 	buf = kmalloc(DOWNLOAD_CHUCK + pad_size, GFP_KERNEL);
···
 		goto out;
 
 out:
-	filp_close(filp, current->files);
+	filp_close(filp, NULL);
 
 restore_fs:
 	set_fs(fs);

drivers/target/target_core_file.c

···
 	struct se_subsystem_dev *se_dev,
 	void *p)
 {
-	char *dev_p = NULL;
 	struct se_device *dev;
 	struct se_dev_limits dev_limits;
 	struct queue_limits *limits;
 	struct fd_dev *fd_dev = p;
 	struct fd_host *fd_host = hba->hba_ptr;
-	mm_segment_t old_fs;
 	struct file *file;
 	struct inode *inode = NULL;
 	int dev_flags = 0, flags, ret = -EINVAL;
 
 	memset(&dev_limits, 0, sizeof(struct se_dev_limits));
 
-	old_fs = get_fs();
-	set_fs(get_ds());
-	dev_p = getname(fd_dev->fd_dev_name);
-	set_fs(old_fs);
-
-	if (IS_ERR(dev_p)) {
-		pr_err("getname(%s) failed: %lu\n",
-			fd_dev->fd_dev_name, IS_ERR(dev_p));
-		ret = PTR_ERR(dev_p);
-		goto fail;
-	}
 	/*
 	 * Use O_DSYNC by default instead of O_SYNC to forgo syncing
 	 * of pure timestamp updates.
 	 */
 	flags = O_RDWR | O_CREAT | O_LARGEFILE | O_DSYNC;
 
-	file = filp_open(dev_p, flags, 0600);
+	file = filp_open(fd_dev->fd_dev_name, flags, 0600);
 	if (IS_ERR(file)) {
-		pr_err("filp_open(%s) failed\n", dev_p);
+		pr_err("filp_open(%s) failed\n", fd_dev->fd_dev_name);
 		ret = PTR_ERR(file);
-		goto fail;
-	}
-	if (!file || !file->f_dentry) {
-		pr_err("filp_open(%s) failed\n", dev_p);
 		goto fail;
 	}
 	fd_dev->fd_file = file;
···
 		" %llu total bytes\n", fd_host->fd_host_id, fd_dev->fd_dev_id,
 			fd_dev->fd_dev_name, fd_dev->fd_dev_size);
 
-	putname(dev_p);
 	return dev;
 fail:
 	if (fd_dev->fd_file) {
 		filp_close(fd_dev->fd_file, NULL);
 		fd_dev->fd_file = NULL;
 	}
-	putname(dev_p);
 	return ERR_PTR(ret);
 }
 
···
 		token = match_token(ptr, tokens, args);
 		switch (token) {
 		case Opt_fd_dev_name:
-			arg_p = match_strdup(&args[0]);
-			if (!arg_p) {
-				ret = -ENOMEM;
+			if (match_strlcpy(fd_dev->fd_dev_name, &args[0],
+				FD_MAX_DEV_NAME) == 0) {
+				ret = -EINVAL;
 				break;
 			}
-			snprintf(fd_dev->fd_dev_name, FD_MAX_DEV_NAME,
-					"%s", arg_p);
-			kfree(arg_p);
 			pr_debug("FILEIO: Referencing Path: %s\n",
 					fd_dev->fd_dev_name);
 			fd_dev->fbd_flags |= FBDF_HAS_PATH;

drivers/usb/gadget/storage_common.c

···
 	if (!(filp->f_mode & FMODE_WRITE))
 		ro = 1;
 
-	if (filp->f_path.dentry)
-		inode = filp->f_path.dentry->d_inode;
-	if (!inode || (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))) {
+	inode = filp->f_path.dentry->d_inode;
+	if ((!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))) {
 		LINFO(curlun, "invalid file type: %s\n", filename);
 		goto out;
 	}
···
 	 * If we can't read the file, it's no good.
 	 * If we can't write the file, use it read-only.
 	 */
-	if (!filp->f_op || !(filp->f_op->read || filp->f_op->aio_read)) {
+	if (!(filp->f_op->read || filp->f_op->aio_read)) {
 		LINFO(curlun, "file not readable: %s\n", filename);
 		goto out;
 	}
···
 	if (fsg_lun_is_open(curlun))
 		fsg_lun_close(curlun);
 
-	get_file(filp);
 	curlun->blksize = blksize;
 	curlun->blkbits = blkbits;
 	curlun->ro = ro;
···
 	curlun->file_length = size;
 	curlun->num_sectors = num_sectors;
 	LDBG(curlun, "open backing file: %s\n", filename);
-	rc = 0;
+	return 0;
 
 out:
-	filp_close(filp, current->files);
+	fput(filp);
 	return rc;
 }
 

drivers/usb/gadget/u_uac1.c

···
 	/* Close control device */
 	snd = &gau->control;
 	if (snd->filp)
-		filp_close(snd->filp, current->files);
+		filp_close(snd->filp, NULL);
 
 	/* Close PCM playback device and setup substream */
 	snd = &gau->playback;
 	if (snd->filp)
-		filp_close(snd->filp, current->files);
+		filp_close(snd->filp, NULL);
 
 	/* Close PCM capture device and setup substream */
 	snd = &gau->capture;
 	if (snd->filp)
-		filp_close(snd->filp, current->files);
+		filp_close(snd->filp, NULL);
 
 	return 0;
 }

drivers/video/fb_defio.c

···
 	deferred framebuffer IO. then if userspace touches a page
 	again, we repeat the same scheme */
 
+	file_update_time(vma->vm_file);
+
 	/* protect against the workqueue changing the page list */
 	mutex_lock(&fbdefio->lock);
 

fs/9p/vfs_file.c

···
 	p9_debug(P9_DEBUG_VFS, "page %p fid %lx\n",
 		 page, (unsigned long)filp->private_data);
 
+	/* Update file times before taking page lock */
+	file_update_time(filp);
+
 	v9inode = V9FS_I(inode);
 	/* make sure the cache has finished storing the page */
 	v9fs_fscache_wait_on_page_write(inode, page);

fs/btrfs/disk-io.c

···
 	struct btrfs_root *root = arg;
 
 	do {
-		vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
-
 		if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
 		    mutex_trylock(&root->fs_info->cleaner_mutex)) {
 			btrfs_run_delayed_iputs(root);
···
 	do {
 		cannot_commit = false;
 		delay = HZ * 30;
-		vfs_check_frozen(root->fs_info->sb, SB_FREEZE_WRITE);
 		mutex_lock(&root->fs_info->transaction_kthread_mutex);
 
 		spin_lock(&root->fs_info->trans_lock);

fs/btrfs/file.c

···
 	ssize_t err = 0;
 	size_t count, ocount;
 
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+	sb_start_write(inode->i_sb);
 
 	mutex_lock(&inode->i_mutex);
 
···
 			num_written = err;
 	}
 out:
+	sb_end_write(inode->i_sb);
 	current->backing_dev_info = NULL;
 	return num_written ? num_written : err;
 }

fs/btrfs/inode.c

···
 	u64 page_start;
 	u64 page_end;
 
+	sb_start_pagefault(inode->i_sb);
 	ret  = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE);
 	if (!ret) {
 		ret = file_update_time(vma->vm_file);
···
 	unlock_extent_cached(io_tree, page_start, page_end, &cached_state, GFP_NOFS);
 
 out_unlock:
-	if (!ret)
+	if (!ret) {
+		sb_end_pagefault(inode->i_sb);
 		return VM_FAULT_LOCKED;
+	}
 	unlock_page(page);
 out:
 	btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE);
 out_noreserve:
+	sb_end_pagefault(inode->i_sb);
 	return ret;
 }
 

fs/btrfs/ioctl.c

···
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
+	ret = mnt_want_write_file(file);
+	if (ret)
+		return ret;
+
 	mutex_lock(&inode->i_mutex);
 
 	ip_oldflags = ip->flags;
···
 			goto out_unlock;
 		}
 	}
-
-	ret = mnt_want_write_file(file);
-	if (ret)
-		goto out_unlock;
 
 	if (flags & FS_SYNC_FL)
 		ip->flags |= BTRFS_INODE_SYNC;
···
 		inode->i_flags = i_oldflags;
 	}
 
-	mnt_drop_write_file(file);
  out_unlock:
 	mutex_unlock(&inode->i_mutex);
+	mnt_drop_write_file(file);
 	return ret;
 }
 
···
 	struct dentry *dentry;
 	int error;
 
+	error = mnt_want_write(parent->mnt);
+	if (error)
+		return error;
+
 	mutex_lock_nested(&dir->i_mutex, I_MUTEX_PARENT);
 
 	dentry = lookup_one_len(name, parent->dentry, namelen);
···
 	dput(dentry);
 out_unlock:
 	mutex_unlock(&dir->i_mutex);
+	mnt_drop_write(parent->mnt);
 	return error;
 }
 

fs/btrfs/transaction.c

···
 	if (!h)
 		return ERR_PTR(-ENOMEM);
 
+	sb_start_intwrite(root->fs_info->sb);
+
 	if (may_wait_transaction(root, type))
 		wait_current_trans(root);
 
···
 	} while (ret == -EBUSY);
 
 	if (ret < 0) {
+		sb_end_intwrite(root->fs_info->sb);
 		kmem_cache_free(btrfs_trans_handle_cachep, h);
 		return ERR_PTR(ret);
 	}
···
 	}
 	btrfs_trans_release_metadata(trans, root);
 	trans->block_rsv = NULL;
+
+	sb_end_intwrite(root->fs_info->sb);
 
 	if (lock && !atomic_read(&root->fs_info->open_ioctl_trans) &&
 	    should_end_transaction(trans, root)) {
···
 
 	put_transaction(cur_trans);
 	put_transaction(cur_trans);
+
+	sb_end_intwrite(root->fs_info->sb);
 
 	trace_btrfs_transaction_commit(root);
 

fs/buffer.c

···
  * beyond EOF, then the page is guaranteed safe against truncation until we
  * unlock the page.
  *
- * Direct callers of this function should call vfs_check_frozen() so that page
- * fault does not busyloop until the fs is thawed.
+ * Direct callers of this function should protect against filesystem freezing
+ * using sb_start_write() - sb_end_write() functions.
  */
 int __block_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf,
 			 get_block_t get_block)
···
 	unsigned long end;
 	loff_t size;
 	int ret;
+
+	/*
+	 * Update file times before taking page lock. We may end up failing the
+	 * fault so this update may be superfluous but who really cares...
+	 */
+	file_update_time(vma->vm_file);
 
 	lock_page(page);
 	size = i_size_read(inode);
···
 
 	if (unlikely(ret < 0))
 		goto out_unlock;
-	/*
-	 * Freezing in progress? We check after the page is marked dirty and
-	 * with page lock held so if the test here fails, we are sure freezing
-	 * code will wait during syncing until the page fault is done - at that
-	 * point page will be dirty and unlocked so freezing code will write it
-	 * and writeprotect it again.
-	 */
 	set_page_dirty(page);
-	if (inode->i_sb->s_frozen != SB_UNFROZEN) {
-		ret = -EAGAIN;
-		goto out_unlock;
-	}
 	wait_on_page_writeback(page);
 	return 0;
 out_unlock:
···
 	int ret;
 	struct super_block *sb = vma->vm_file->f_path.dentry->d_inode->i_sb;
 
-	/*
-	 * This check is racy but catches the common case. The check in
-	 * __block_page_mkwrite() is reliable.
-	 */
-	vfs_check_frozen(sb, SB_FREEZE_WRITE);
+	sb_start_pagefault(sb);
 	ret = __block_page_mkwrite(vma, vmf, get_block);
+	sb_end_pagefault(sb);
 	return block_page_mkwrite_return(ret);
 }
 EXPORT_SYMBOL(block_page_mkwrite);

fs/ceph/addr.c

···
 	loff_t size, len;
 	int ret;
 
+	/* Update time before taking page lock */
+	file_update_time(vma->vm_file);
+
 	size = i_size_read(inode);
 	if (off + PAGE_CACHE_SIZE <= size)
 		len = PAGE_CACHE_SIZE;

fs/ecryptfs/inode.c

···
 	struct vfsmount *lower_mnt;
 	int rc = 0;
 
-	lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent));
-	fsstack_copy_attr_atime(dir_inode, lower_dentry->d_parent->d_inode);
-	BUG_ON(!lower_dentry->d_count);
-
 	dentry_info = kmem_cache_alloc(ecryptfs_dentry_info_cache, GFP_KERNEL);
-	ecryptfs_set_dentry_private(dentry, dentry_info);
 	if (!dentry_info) {
 		printk(KERN_ERR "%s: Out of memory whilst attempting "
 		       "to allocate ecryptfs_dentry_info struct\n",
 			__func__);
 		dput(lower_dentry);
-		mntput(lower_mnt);
-		d_drop(dentry);
 		return -ENOMEM;
 	}
+
+	lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent));
+	fsstack_copy_attr_atime(dir_inode, lower_dentry->d_parent->d_inode);
+	BUG_ON(!lower_dentry->d_count);
+
+	ecryptfs_set_dentry_private(dentry, dentry_info);
 	ecryptfs_set_dentry_lower(dentry, lower_dentry);
 	ecryptfs_set_dentry_lower_mnt(dentry, lower_mnt);
 
···
 	struct dentry *lower_dir_dentry, *lower_dentry;
 	int rc = 0;
 
-	if ((ecryptfs_dentry->d_name.len == 1
-	     && !strcmp(ecryptfs_dentry->d_name.name, "."))
-	    || (ecryptfs_dentry->d_name.len == 2
-		&& !strcmp(ecryptfs_dentry->d_name.name, ".."))) {
-		goto out_d_drop;
-	}
 	lower_dir_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry->d_parent);
 	mutex_lock(&lower_dir_dentry->d_inode->i_mutex);
 	lower_dentry = lookup_one_len(ecryptfs_dentry->d_name.name,
···
 		rc = PTR_ERR(lower_dentry);
 		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
 				"[%d] on lower_dentry = [%s]\n", __func__, rc,
-				encrypted_and_encoded_name);
-		goto out_d_drop;
+				ecryptfs_dentry->d_name.name);
+		goto out;
 	}
 	if (lower_dentry->d_inode)
 		goto interpose;
···
 	if (rc) {
 		printk(KERN_ERR "%s: Error attempting to encrypt and encode "
 		       "filename; rc = [%d]\n", __func__, rc);
-		goto out_d_drop;
+		goto out;
 	}
 	mutex_lock(&lower_dir_dentry->d_inode->i_mutex);
 	lower_dentry = lookup_one_len(encrypted_and_encoded_name,
···
 		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
 				"[%d] on lower_dentry = [%s]\n", __func__, rc,
 				encrypted_and_encoded_name);
-		goto out_d_drop;
+		goto out;
 	}
 interpose:
 	rc = ecryptfs_lookup_interpose(ecryptfs_dentry, lower_dentry,
 				       ecryptfs_dir_inode);
-	goto out;
-out_d_drop:
-	d_drop(ecryptfs_dentry);
 out:
 	kfree(encrypted_and_encoded_name);
 	return ERR_PTR(rc);

fs/exec.c

···
  */
 static int umh_pipe_setup(struct subprocess_info *info, struct cred *new)
 {
-	struct file *rp, *wp;
+	struct file *files[2];
 	struct fdtable *fdt;
 	struct coredump_params *cp = (struct coredump_params *)info->data;
 	struct files_struct *cf = current->files;
+	int err = create_pipe_files(files, 0);
+	if (err)
+		return err;
 
-	wp = create_write_pipe(0);
-	if (IS_ERR(wp))
-		return PTR_ERR(wp);
-
-	rp = create_read_pipe(wp, 0);
-	if (IS_ERR(rp)) {
-		free_write_pipe(wp);
-		return PTR_ERR(rp);
-	}
-
-	cp->file = wp;
+	cp->file = files[1];
 
 	sys_close(0);
-	fd_install(0, rp);
+	fd_install(0, files[0]);
 	spin_lock(&cf->file_lock);
 	fdt = files_fdtable(cf);
 	__set_open_fd(0, fdt);

fs/ext2/inode.c

···
 	truncate_inode_pages(&inode->i_data, 0);
 
 	if (want_delete) {
+		sb_start_intwrite(inode->i_sb);
 		/* set dtime */
 		EXT2_I(inode)->i_dtime	= get_seconds();
 		mark_inode_dirty(inode);
···
 	if (unlikely(rsv))
 		kfree(rsv);
 
-	if (want_delete)
+	if (want_delete) {
 		ext2_free_inode(inode);
+		sb_end_intwrite(inode->i_sb);
+	}
 }
 
 typedef struct {

fs/ext2/super.c

···
 static int ext2_remount (struct super_block * sb, int * flags, char * data);
 static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf);
 static int ext2_sync_fs(struct super_block *sb, int wait);
+static int ext2_freeze(struct super_block *sb);
+static int ext2_unfreeze(struct super_block *sb);
 
 void ext2_error(struct super_block *sb, const char *function,
 		const char *fmt, ...)
···
 	.evict_inode	= ext2_evict_inode,
 	.put_super	= ext2_put_super,
 	.sync_fs	= ext2_sync_fs,
+	.freeze_fs	= ext2_freeze,
+	.unfreeze_fs	= ext2_unfreeze,
 	.statfs		= ext2_statfs,
 	.remount_fs	= ext2_remount,
 	.show_options	= ext2_show_options,
···
 	return 0;
 }
 
+static int ext2_freeze(struct super_block *sb)
+{
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
+
+	/*
+	 * Open but unlinked files present? Keep EXT2_VALID_FS flag cleared
+	 * because we have unattached inodes and thus filesystem is not fully
+	 * consistent.
+	 */
+	if (atomic_long_read(&sb->s_remove_count)) {
+		ext2_sync_fs(sb, 1);
+		return 0;
+	}
+	/* Set EXT2_FS_VALID flag */
+	spin_lock(&sbi->s_lock);
+	sbi->s_es->s_state = cpu_to_le16(sbi->s_mount_state);
+	spin_unlock(&sbi->s_lock);
+	ext2_sync_super(sb, sbi->s_es, 1);
+
+	return 0;
+}
+
+static int ext2_unfreeze(struct super_block *sb)
+{
+	/* Just write sb to clear EXT2_VALID_FS flag */
+	ext2_write_super(sb);
+
+	return 0;
+}
 
 void ext2_write_super(struct super_block *sb)
 {

fs/ext4/inode.c

···
 	if (is_bad_inode(inode))
 		goto no_delete;
 
+	/*
+	 * Protect us against freezing - iput() caller didn't have to have any
+	 * protection against it
+	 */
+	sb_start_intwrite(inode->i_sb);
 	handle = ext4_journal_start(inode, ext4_blocks_for_truncate(inode)+3);
 	if (IS_ERR(handle)) {
 		ext4_std_error(inode->i_sb, PTR_ERR(handle));
···
 		 * cleaned up.
 		 */
 		ext4_orphan_del(NULL, inode);
+		sb_end_intwrite(inode->i_sb);
 		goto no_delete;
 	}
 
···
 		stop_handle:
 			ext4_journal_stop(handle);
 			ext4_orphan_del(NULL, inode);
+			sb_end_intwrite(inode->i_sb);
 			goto no_delete;
 		}
 	}
···
 	else
 		ext4_free_inode(handle, inode);
 	ext4_journal_stop(handle);
+	sb_end_intwrite(inode->i_sb);
 	return;
 no_delete:
 	ext4_clear_inode(inode);	/* We must guarantee clearing of inode... */
···
 	get_block_t *get_block;
 	int retries = 0;
 
-	/*
-	 * This check is racy but catches the common case. We rely on
-	 * __block_page_mkwrite() to do a reliable check.
-	 */
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+	sb_start_pagefault(inode->i_sb);
 	/* Delalloc case is easy... */
 	if (test_opt(inode->i_sb, DELALLOC) &&
 	    !ext4_should_journal_data(inode) &&
···
 out_ret:
 	ret = block_page_mkwrite_return(ret);
 out:
+	sb_end_pagefault(inode->i_sb);
 	return ret;
 }

fs/ext4/mmp.c

···
 {
 	struct mmp_struct *mmp = (struct mmp_struct *)(bh->b_data);
 
+	/*
+	 * We protect against freezing so that we don't create dirty buffers
+	 * on frozen filesystem.
+	 */
+	sb_start_write(sb);
 	ext4_mmp_csum_set(sb, mmp);
 	mark_buffer_dirty(bh);
 	lock_buffer(bh);
···
 	get_bh(bh);
 	submit_bh(WRITE_SYNC, bh);
 	wait_on_buffer(bh);
+	sb_end_write(sb);
 	if (unlikely(!buffer_uptodate(bh)))
 		return 1;
 

fs/ext4/super.c

···
  * journal_end calls result in the superblock being marked dirty, so
  * that sync() will call the filesystem's write_super callback if
  * appropriate.
- *
- * To avoid j_barrier hold in userspace when a user calls freeze(),
- * ext4 prevents a new handle from being started by s_frozen, which
- * is in an upper layer.
  */
 handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks)
 {
 	journal_t *journal;
-	handle_t  *handle;
 
 	trace_ext4_journal_start(sb, nblocks, _RET_IP_);
 	if (sb->s_flags & MS_RDONLY)
 		return ERR_PTR(-EROFS);
 
+	WARN_ON(sb->s_writers.frozen == SB_FREEZE_COMPLETE);
 	journal = EXT4_SB(sb)->s_journal;
-	handle = ext4_journal_current_handle();
-
-	/*
-	 * If a handle has been started, it should be allowed to
-	 * finish, otherwise deadlock could happen between freeze
-	 * and others(e.g. truncate) due to the restart of the
-	 * journal handle if the filesystem is forzen and active
-	 * handles are not stopped.
-	 */
-	if (!handle)
-		vfs_check_frozen(sb, SB_FREEZE_TRANS);
-
 	if (!journal)
 		return ext4_get_nojournal();
 	/*
···
 	sb = elr->lr_super;
 	ngroups = EXT4_SB(sb)->s_groups_count;
 
+	sb_start_write(sb);
 	for (group = elr->lr_next_group; group < ngroups; group++) {
 		gdp = ext4_get_group_desc(sb, group, NULL);
 		if (!gdp) {
···
 		elr->lr_next_sched = jiffies + elr->lr_timeout;
 		elr->lr_next_group = group + 1;
 	}
+	sb_end_write(sb);
 
 	return ret;
 }
···
 		return 0;
 
 	journal = EXT4_SB(sb)->s_journal;
-	if (journal) {
-		vfs_check_frozen(sb, SB_FREEZE_TRANS);
+	if (journal)
 		ret = ext4_journal_force_commit(journal);
-	}
 
 	return ret;
 }
···
  * gives us a chance to flush the journal completely and mark the fs clean.
  *
  * Note that only this function cannot bring a filesystem to be in a clean
- * state independently, because ext4 prevents a new handle from being started
- * by @sb->s_frozen, which stays in an upper layer.  It thus needs help from
- * the upper layer.
+ * state independently. It relies on upper layer to stop all data & metadata
+ * modifications.
  */
 static int ext4_freeze(struct super_block *sb)
 {
···
 	EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
 	error = ext4_commit_super(sb, 1);
 out:
-	/* we rely on s_frozen to stop further updates */
+	/* we rely on upper layer to stop further updates */
 	jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
 	return error;
 }

fs/fat/file.c

···
 	if (err)
 		goto out;
 
-	mutex_lock(&inode->i_mutex);
 	err = mnt_want_write_file(file);
 	if (err)
-		goto out_unlock_inode;
+		goto out;
+	mutex_lock(&inode->i_mutex);
 
 	/*
 	 * ATTR_VOLUME and ATTR_DIR cannot be changed; this also
···
 	/* The root directory has no attributes */
 	if (inode->i_ino == MSDOS_ROOT_INO && attr != ATTR_DIR) {
 		err = -EINVAL;
-		goto out_drop_write;
+		goto out_unlock_inode;
 	}
 
 	if (sbi->options.sys_immutable &&
 	    ((attr | oldattr) & ATTR_SYS) &&
 	    !capable(CAP_LINUX_IMMUTABLE)) {
 		err = -EPERM;
-		goto out_drop_write;
+		goto out_unlock_inode;
 	}
 
 	/*
···
 	 */
 	err = security_inode_setattr(file->f_path.dentry, &ia);
 	if (err)
-		goto out_drop_write;
+		goto out_unlock_inode;
 
 	/* This MUST be done before doing anything irreversible... */
 	err = fat_setattr(file->f_path.dentry, &ia);
 	if (err)
-		goto out_drop_write;
+		goto out_unlock_inode;
 
 	fsnotify_change(file->f_path.dentry, ia.ia_valid);
 	if (sbi->options.sys_immutable) {
···
 
 	fat_save_attrs(inode, attr);
 	mark_inode_dirty(inode);
-out_drop_write:
-	mnt_drop_write_file(file);
 out_unlock_inode:
 	mutex_unlock(&inode->i_mutex);
+	mnt_drop_write_file(file);
 out:
 	return err;
 }

fs/file_table.c

···
 
 static struct percpu_counter nr_files __cacheline_aligned_in_smp;
 
-static inline void file_free_rcu(struct rcu_head *head)
+static void file_free_rcu(struct rcu_head *head)
 {
 	struct file *f = container_of(head, struct file, f_u.fu_rcuhead);
 
···
 		return;
 	if (file_check_writeable(file) != 0)
 		return;
-	mnt_drop_write(mnt);
+	__mnt_drop_write(mnt);
 	file_release_write(file);
 }
 

fs/fuse/file.c

···
 		return err;
 
 	count = ocount;
-
+	sb_start_write(inode->i_sb);
 	mutex_lock(&inode->i_mutex);
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
 
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = mapping->backing_dev_info;
···
 out:
 	current->backing_dev_info = NULL;
 	mutex_unlock(&inode->i_mutex);
+	sb_end_write(inode->i_sb);
 
 	return written ? written : err;
 }

fs/gfs2/file.c

···
 	loff_t size;
 	int ret;
 
-	/* Wait if fs is frozen. This is racy so we check again later on
-	 * and retry if the fs has been frozen after the page lock has
-	 * been acquired
-	 */
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+	sb_start_pagefault(inode->i_sb);
+
+	/* Update file times before taking page lock */
+	file_update_time(vma->vm_file);
 
 	ret = gfs2_rs_alloc(ip);
 	if (ret)
···
 	gfs2_holder_uninit(&gh);
 	if (ret == 0) {
 		set_page_dirty(page);
-		/* This check must be post dropping of transaction lock */
-		if (inode->i_sb->s_frozen == SB_UNFROZEN) {
-			wait_on_page_writeback(page);
-		} else {
-			ret = -EAGAIN;
-			unlock_page(page);
-		}
+		wait_on_page_writeback(page);
 	}
+	sb_end_pagefault(inode->i_sb);
 	return block_page_mkwrite_return(ret);
 }
 

fs/gfs2/trans.c

···
 	if (revokes)
 		tr->tr_reserved += gfs2_struct2blk(sdp, revokes,
 						   sizeof(u64));
+	sb_start_intwrite(sdp->sd_vfs);
 	gfs2_holder_init(sdp->sd_trans_gl, LM_ST_SHARED, 0, &tr->tr_t_gh);
 
 	error = gfs2_glock_nq(&tr->tr_t_gh);
···
 	gfs2_glock_dq(&tr->tr_t_gh);
 
 fail_holder_uninit:
+	sb_end_intwrite(sdp->sd_vfs);
 	gfs2_holder_uninit(&tr->tr_t_gh);
 	kfree(tr);
 
···
 			gfs2_holder_uninit(&tr->tr_t_gh);
 			kfree(tr);
 		}
+		sb_end_intwrite(sdp->sd_vfs);
 		return;
 	}
 
···
 
 	if (sdp->sd_vfs->s_flags & MS_SYNCHRONOUS)
 		gfs2_log_flush(sdp, NULL);
+	sb_end_intwrite(sdp->sd_vfs);
 }
 
 /**

fs/inode.c

···
 	if (timespec_equal(&inode->i_atime, &now))
 		return;
 
-	if (mnt_want_write(mnt))
+	if (!sb_start_write_trylock(inode->i_sb))
 		return;
 
+	if (__mnt_want_write(mnt))
+		goto skip_update;
 	/*
 	 * File systems can error out when updating inodes if they need to
 	 * allocate new space to modify an inode (such is the case for
···
 	 * of the fs read only, e.g. subvolumes in Btrfs.
 	 */
 	update_time(inode, &now, S_ATIME);
-	mnt_drop_write(mnt);
+	__mnt_drop_write(mnt);
+skip_update:
+	sb_end_write(inode->i_sb);
 }
 EXPORT_SYMBOL(touch_atime);
 
···
 		return 0;
 
 	/* Finally allowed to write? Takes lock. */
-	if (mnt_want_write_file(file))
+	if (__mnt_want_write_file(file))
 		return 0;
 
 	ret = update_time(inode, &now, sync_it);
-	mnt_drop_write_file(file);
+	__mnt_drop_write_file(file);
 
 	return ret;
 }

fs/internal.h

···
 
 extern struct lglock vfsmount_lock;
 
+extern int __mnt_want_write(struct vfsmount *);
+extern int __mnt_want_write_file(struct file *);
+extern void __mnt_drop_write(struct vfsmount *);
+extern void __mnt_drop_write_file(struct file *);
 
 /*
  * fs_struct.c

fs/lockd/clntproc.c

···
 	struct nlm_rqst		*call;
 	int			status;
 
-	nlm_get_host(host);
 	call = nlm_alloc_call(host);
 	if (call == NULL)
 		return -ENOMEM;
 
 	nlmclnt_locks_init_private(fl, host);
+	if (!fl->fl_u.nfs_fl.owner) {
+		/* lockowner allocation has failed */
+		nlmclnt_release_call(call);
+		return -ENOMEM;
+	}
 	/* Set up the argument struct */
 	nlmclnt_setlockargs(call, fl);
 
···
 
 /*
  * Allocate an NLM RPC call struct
- *
- * Note: the caller must hold a reference to host. In case of failure,
- * this reference will be released.
  */
 struct nlm_rqst *nlm_alloc_call(struct nlm_host *host)
 {
···
 			atomic_set(&call->a_count, 1);
 			locks_init_lock(&call->a_args.lock.fl);
 			locks_init_lock(&call->a_res.lock.fl);
-			call->a_host = host;
+			call->a_host = nlm_get_host(host);
 			return call;
 		}
 		if (signalled())
···
 		printk("nlm_alloc_call: failed, waiting for memory\n");
 		schedule_timeout_interruptible(5*HZ);
 	}
-	nlmclnt_release_host(host);
 	return NULL;
 }
 
···
 	dprintk("lockd: blocking lock attempt was interrupted by a signal.\n"
 		"       Attempting to cancel lock.\n");
 
-	req = nlm_alloc_call(nlm_get_host(host));
+	req = nlm_alloc_call(host);
 	if (!req)
 		return -ENOMEM;
 	req->a_flags = RPC_TASK_ASYNC;

fs/lockd/svc4proc.c

···
 		return rpc_system_err;
 
 	call = nlm_alloc_call(host);
+	nlmsvc_release_host(host);
 	if (call == NULL)
 		return rpc_system_err;
 

fs/lockd/svclock.c

···
 	struct nlm_block	*block;
 	struct nlm_rqst		*call = NULL;
 
-	nlm_get_host(host);
 	call = nlm_alloc_call(host);
 	if (call == NULL)
 		return NULL;

fs/lockd/svcproc.c

···
 		return rpc_system_err;
 
 	call = nlm_alloc_call(host);
+	nlmsvc_release_host(host);
 	if (call == NULL)
 		return rpc_system_err;
 

fs/namei.c

···
 	path_put(link);
 }
 
+int sysctl_protected_symlinks __read_mostly = 1;
+int sysctl_protected_hardlinks __read_mostly = 1;
+
+/**
+ * may_follow_link - Check symlink following for unsafe situations
+ * @link: The path of the symlink
+ *
+ * In the case of the sysctl_protected_symlinks sysctl being enabled,
+ * CAP_DAC_OVERRIDE needs to be specifically ignored if the symlink is
+ * in a sticky world-writable directory. This is to protect privileged
+ * processes from failing races against path names that may change out
+ * from under them by way of other users creating malicious symlinks.
+ * It will permit symlinks to be followed only when outside a sticky
+ * world-writable directory, or when the uid of the symlink and follower
+ * match, or when the directory owner matches the symlink's owner.
+ *
+ * Returns 0 if following the symlink is allowed, -ve on error.
+ */
+static inline int may_follow_link(struct path *link, struct nameidata *nd)
+{
+	const struct inode *inode;
+	const struct inode *parent;
+
+	if (!sysctl_protected_symlinks)
+		return 0;
+
+	/* Allowed if owner and follower match. */
+	inode = link->dentry->d_inode;
+	if (current_cred()->fsuid == inode->i_uid)
+		return 0;
+
+	/* Allowed if parent directory not sticky and world-writable. */
+	parent = nd->path.dentry->d_inode;
+	if ((parent->i_mode & (S_ISVTX|S_IWOTH)) != (S_ISVTX|S_IWOTH))
+		return 0;
+
+	/* Allowed if parent directory and link owner match. */
+	if (parent->i_uid == inode->i_uid)
+		return 0;
+
+	path_put_conditional(link, nd);
+	path_put(&nd->path);
+	audit_log_link_denied("follow_link", link);
+	return -EACCES;
+}
+
+/**
+ * safe_hardlink_source - Check for safe hardlink conditions
+ * @inode: the source inode to hardlink from
+ *
+ * Return false if at least one of the following conditions:
+ *    - inode is not a regular file
+ *    - inode is setuid
+ *    - inode is setgid and group-exec
+ *    - access failure for read and write
+ *
+ * Otherwise returns true.
+ */
+static bool safe_hardlink_source(struct inode *inode)
+{
+	umode_t mode = inode->i_mode;
+
+	/* Special files should not get pinned to the filesystem. */
+	if (!S_ISREG(mode))
+		return false;
+
+	/* Setuid files should not get pinned to the filesystem. */
+	if (mode & S_ISUID)
+		return false;
+
+	/* Executable setgid files should not get pinned to the filesystem. */
+	if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP))
+		return false;
+
+	/* Hardlinking to unreadable or unwritable sources is dangerous. */
+	if (inode_permission(inode, MAY_READ | MAY_WRITE))
+		return false;
+
+	return true;
+}
+
+/**
+ * may_linkat - Check permissions for creating a hardlink
+ * @link: the source to hardlink from
+ *
+ * Block hardlink when all of:
+ *  - sysctl_protected_hardlinks enabled
+ *  - fsuid does not match inode
+ *  - hardlink source is unsafe (see safe_hardlink_source() above)
+ *  - not CAP_FOWNER
+ *
+ * Returns 0 if successful, -ve on error.
+ */
+static int may_linkat(struct path *link)
+{
+	const struct cred *cred;
+	struct inode *inode;
+
+	if (!sysctl_protected_hardlinks)
+		return 0;
+
+	cred = current_cred();
+	inode = link->dentry->d_inode;
+
+	/* Source inode owner (or CAP_FOWNER) can hardlink all they like,
+	 * otherwise, it must be a safe source.
+	 */
+	if (cred->fsuid == inode->i_uid || safe_hardlink_source(inode) ||
+	    capable(CAP_FOWNER))
+		return 0;
+
+	audit_log_link_denied("linkat", link);
+	return -EPERM;
+}
+
 static __always_inline int
 follow_link(struct path *link, struct nameidata *nd, void **p)
 {
···
 		while (err > 0) {
 			void *cookie;
 			struct path link = path;
+			err = may_follow_link(&link, nd);
+			if (unlikely(err))
+				break;
 			nd->flags |= LOOKUP_PARENT;
 			err = follow_link(&link, nd, &cookie);
 			if (err)
···
 static int atomic_open(struct nameidata *nd, struct dentry *dentry,
 			struct path *path, struct file *file,
 			const struct open_flags *op,
-			bool *want_write, bool need_lookup,
+			bool got_write, bool need_lookup,
 			int *opened)
 {
 	struct inode *dir =  nd->path.dentry->d_inode;
···
 	if ((open_flag & O_CREAT) && !IS_POSIXACL(dir))
 		mode &= ~current_umask();
 
-	if (open_flag & O_EXCL) {
+	if ((open_flag & (O_EXCL | O_CREAT)) == (O_EXCL | O_CREAT)) {
 		open_flag &= ~O_TRUNC;
 		*opened |= FILE_CREATED;
 	}
···
 	 * Another problem is returing the "right" error value (e.g. for an
 	 * O_EXCL open we want to return EEXIST not EROFS).
 	 */
-	if ((open_flag & (O_CREAT | O_TRUNC)) ||
-	    (open_flag & O_ACCMODE) != O_RDONLY) {
-		error = mnt_want_write(nd->path.mnt);
-		if (!error) {
-			*want_write = true;
-		} else if (!(open_flag & O_CREAT)) {
+	if (((open_flag & (O_CREAT | O_TRUNC)) ||
+	    (open_flag & O_ACCMODE) != O_RDONLY) && unlikely(!got_write)) {
+		if (!(open_flag & O_CREAT)) {
 			/*
 			 * No O_CREATE -> atomicity not a requirement -> fall
 			 * back to lookup + open
···
 			goto no_open;
 		} else if (open_flag & (O_EXCL | O_TRUNC)) {
 			/* Fall back and fail with the right error */
-			create_error = error;
+			create_error = -EROFS;
 			goto no_open;
 		} else {
 			/* No side effects, safe to clear O_CREAT */
-			create_error = error;
+			create_error = -EROFS;
 			open_flag &= ~O_CREAT;
 		}
 	}
···
 static int lookup_open(struct nameidata *nd, struct path *path,
 			struct file *file,
 			const struct open_flags *op,
-			bool *want_write, int *opened)
+			bool got_write, int *opened)
 {
 	struct dentry *dir = nd->path.dentry;
 	struct inode *dir_inode = dir->d_inode;
···
 		goto out_no_open;
 
 	if ((nd->flags & LOOKUP_OPEN) && dir_inode->i_op->atomic_open) {
-		return atomic_open(nd, dentry, path, file, op, want_write,
+		return atomic_open(nd, dentry, path, file, op, got_write,
 				   need_lookup, opened);
 	}
 
···
 		 * a permanent write count is taken through
 		 * the 'struct file' in finish_open().
 		 */
-		error = mnt_want_write(nd->path.mnt);
-		if (error)
+		if (!got_write) {
+			error = -EROFS;
 			goto out_dput;
-		*want_write = true;
+		}
 		*opened |= FILE_CREATED;
 		error = security_path_mknod(&nd->path, dentry, mode, 0);
 		if (error)
···
 	struct dentry *dir = nd->path.dentry;
 	int open_flag = op->open_flag;
 	bool will_truncate = (open_flag & O_TRUNC) != 0;
-	bool want_write = false;
+	bool got_write = false;
 	int acc_mode = op->acc_mode;
 	struct inode *inode;
 	bool symlink_ok = false;
···
 	}
 
 retry_lookup:
+	if (op->open_flag & (O_CREAT | O_TRUNC | O_WRONLY | O_RDWR)) {
+		error = mnt_want_write(nd->path.mnt);
+		if (!error)
+			got_write = true;
+		/*
+		 * do _not_ fail yet - we might not need that or fail with
+		 * a different error; let lookup_open() decide; we'll be
+		 * dropping this one anyway.
+		 */
+	}
 	mutex_lock(&dir->d_inode->i_mutex);
-	error = lookup_open(nd, path, file, op, &want_write, opened);
+	error = lookup_open(nd, path, file, op, got_write, opened);
 	mutex_unlock(&dir->d_inode->i_mutex);
 
 	if (error <= 0) {
···
 	}
 
 	/*
-	 * It already exists.
+	 * create/update audit record if it already exists.
 	 */
-	audit_inode(pathname, path->dentry);
+	if (path->dentry->d_inode)
+		audit_inode(pathname, path->dentry);
 
 	/*
 	 * If atomic_open() acquired write access it is dropped now due to
 	 * possible mount and symlink following (this might be optimized away if
 	 * necessary...)
 	 */
-	if (want_write) {
+	if (got_write) {
 		mnt_drop_write(nd->path.mnt);
-		want_write = false;
+		got_write = false;
 	}
 
 	error = -EEXIST;
-	if (open_flag & O_EXCL)
+	if ((open_flag & (O_EXCL | O_CREAT)) == (O_EXCL | O_CREAT))
 		goto exit_dput;
 
 	error = follow_managed(path, nd->flags);
···
 		error = mnt_want_write(nd->path.mnt);
 		if (error)
 			goto out;
-		want_write = true;
+		got_write = true;
 	}
 finish_open_created:
 	error = may_open(&nd->path, acc_mode, open_flag);
···
 			goto exit_fput;
 	}
 out:
-	if (want_write)
+	if (got_write)
 		mnt_drop_write(nd->path.mnt);
 	path_put(&save_parent);
 	terminate_walk(nd);
···
 	nd->inode = dir->d_inode;
 	save_parent.mnt = NULL;
 	save_parent.dentry = NULL;
-	if (want_write) {
+	if (got_write) {
 		mnt_drop_write(nd->path.mnt);
-		want_write = false;
+		got_write = false;
 	}
 	retried = true;
 	goto retry_lookup;
···
 			error = -ELOOP;
 			break;
 		}
+		error = may_follow_link(&link, nd);
+		if (unlikely(error))
+			break;
 		nd->flags |= LOOKUP_PARENT;
 		nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
 		error = follow_link(&link, nd, &cookie);
···
 {
 	struct dentry *dentry = ERR_PTR(-EEXIST);
 	struct nameidata nd;
+	int err2;
 	int error = do_path_lookup(dfd, pathname, LOOKUP_PARENT, &nd);
 	if (error)
 		return ERR_PTR(error);
···
 	nd.flags &= ~LOOKUP_PARENT;
 	nd.flags |= LOOKUP_CREATE | LOOKUP_EXCL;
 
+	/* don't fail immediately if it's r/o, at least try to report other errors */
+	err2 = mnt_want_write(nd.path.mnt);
 	/*
 	 * Do the final lookup.
 	 */
 	mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
 	dentry = lookup_hash(&nd);
 	if (IS_ERR(dentry))
-		goto fail;
+		goto unlock;
 
+	error = -EEXIST;
 	if (dentry->d_inode)
-		goto eexist;
+		goto fail;
 	/*
 	 * Special case - lookup gave negative, but... we had foo/bar/
 	 * From the vfs_mknod() POV we just have a negative dentry -
···
 	 * been asking for (non-existent) directory. -ENOENT for you.
 	 */
 	if (unlikely(!is_dir && nd.last.name[nd.last.len])) {
-		dput(dentry);
-		dentry = ERR_PTR(-ENOENT);
+		error = -ENOENT;
+		goto fail;
+	}
+	if (unlikely(err2)) {
+		error = err2;
 		goto fail;
 	}
 	*path = nd.path;
 	return dentry;
-eexist:
-	dput(dentry);
-	dentry = ERR_PTR(-EEXIST);
 fail:
+	dput(dentry);
+	dentry = ERR_PTR(error);
+unlock:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+	if (!err2)
+		mnt_drop_write(nd.path.mnt);
 out:
 	path_put(&nd.path);
 	return dentry;
 }
 EXPORT_SYMBOL(kern_path_create);
+
+void done_path_create(struct path *path, struct dentry *dentry)
+{
+	dput(dentry);
+	mutex_unlock(&path->dentry->d_inode->i_mutex);
+	mnt_drop_write(path->mnt);
+	path_put(path);
+}
+EXPORT_SYMBOL(done_path_create);
 
 struct dentry *user_path_create(int dfd, const char __user *pathname, struct path *path, int is_dir)
 {
···
 	struct path path;
 	int error;
 
-	if (S_ISDIR(mode))
-		return -EPERM;
+	error = may_mknod(mode);
+	if (error)
+		return error;
 
 	dentry = user_path_create(dfd, filename, &path, 0);
 	if (IS_ERR(dentry))
···
 
 	if (!IS_POSIXACL(path.dentry->d_inode))
 		mode &= ~current_umask();
-	error = may_mknod(mode);
-	if (error)
-		goto out_dput;
-	error = mnt_want_write(path.mnt);
-	if (error)
-		goto out_dput;
 	error = security_path_mknod(&path, dentry, mode, dev);
 	if (error)
-		goto out_drop_write;
+		goto out;
 	switch (mode & S_IFMT) {
 		case 0: case S_IFREG:
 			error = vfs_create(path.dentry->d_inode,dentry,mode,true);
···
 			error = vfs_mknod(path.dentry->d_inode,dentry,mode,0);
 			break;
 	}
-out_drop_write:
-	mnt_drop_write(path.mnt);
-out_dput:
-	dput(dentry);
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
-
+out:
+	done_path_create(&path, dentry);
 	return error;
 }
 
···
 
 	if (!IS_POSIXACL(path.dentry->d_inode))
 		mode &= ~current_umask();
-	error = mnt_want_write(path.mnt);
-	if (error)
-		goto out_dput;
 	error = security_path_mkdir(&path, dentry, mode);
-	if (error)
-		goto out_drop_write;
-	error = vfs_mkdir(path.dentry->d_inode, dentry, mode);
-out_drop_write:
-	mnt_drop_write(path.mnt);
-out_dput:
-	dput(dentry);
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
+	if (!error)
+		error = vfs_mkdir(path.dentry->d_inode, dentry, mode);
+	done_path_create(&path, dentry);
 	return error;
 }
 
···
 	}
 
 	nd.flags &= ~LOOKUP_PARENT;
+	error = mnt_want_write(nd.path.mnt);
+	if (error)
+		goto exit1;
 
 	mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
 	dentry = lookup_hash(&nd);
···
 		error = -ENOENT;
 		goto exit3;
 	}
-	error = mnt_want_write(nd.path.mnt);
-	if (error)
-		goto exit3;
 	error = security_path_rmdir(&nd.path, dentry);
 	if (error)
-		goto exit4;
+		goto exit3;
 	error = vfs_rmdir(nd.path.dentry->d_inode, dentry);
-exit4:
-	mnt_drop_write(nd.path.mnt);
 exit3:
 	dput(dentry);
 exit2:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
+	mnt_drop_write(nd.path.mnt);
 exit1:
 	path_put(&nd.path);
 	putname(name);
···
 		goto exit1;
 
 	nd.flags &= ~LOOKUP_PARENT;
+	error = mnt_want_write(nd.path.mnt);
+	if (error)
+		goto exit1;
 
 	mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT);
 	dentry = lookup_hash(&nd);
···
 		if (!inode)
 			goto slashes;
 		ihold(inode);
-		error = mnt_want_write(nd.path.mnt);
-		if (error)
-			goto exit2;
 		error = security_path_unlink(&nd.path, dentry);
 		if (error)
-			goto exit3;
+			goto exit2;
 		error = vfs_unlink(nd.path.dentry->d_inode, dentry);
-exit3:
-		mnt_drop_write(nd.path.mnt);
-	exit2:
+exit2:
 		dput(dentry);
 	}
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 	if (inode)
 		iput(inode);	/* truncate the inode here */
+	mnt_drop_write(nd.path.mnt);
 exit1:
 	path_put(&nd.path);
 	putname(name);
···
 	if (IS_ERR(dentry))
 		goto out_putname;
 
-	error = mnt_want_write(path.mnt);
-	if (error)
-		goto out_dput;
 	error = security_path_symlink(&path, dentry, from);
-	if (error)
-		goto out_drop_write;
-	error = vfs_symlink(path.dentry->d_inode, dentry, from);
-out_drop_write:
-	mnt_drop_write(path.mnt);
-out_dput:
-	dput(dentry);
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
+	if (!error)
+		error = vfs_symlink(path.dentry->d_inode, dentry, from);
+	done_path_create(&path, dentry);
 out_putname:
 	putname(from);
 	return error;
···
 	error = -EXDEV;
 	if (old_path.mnt != new_path.mnt)
 		goto out_dput;
-	error = mnt_want_write(new_path.mnt);
-	if (error)
+	error = may_linkat(&old_path);
+	if (unlikely(error))
 		goto out_dput;
 	error = security_path_link(old_path.dentry, &new_path, new_dentry);
 	if (error)
-		goto out_drop_write;
+		goto out_dput;
 	error = vfs_link(old_path.dentry, new_path.dentry->d_inode, new_dentry);
-out_drop_write:
-	mnt_drop_write(new_path.mnt);
 out_dput:
-	dput(new_dentry);
-	mutex_unlock(&new_path.dentry->d_inode->i_mutex);
-	path_put(&new_path);
+	done_path_create(&new_path, new_dentry);
 out:
 	path_put(&old_path);
 
···
 	if (newnd.last_type != LAST_NORM)
 		goto exit2;
 
+	error = mnt_want_write(oldnd.path.mnt);
+	if (error)
+		goto exit2;
+
 	oldnd.flags &= ~LOOKUP_PARENT;
 	newnd.flags &= ~LOOKUP_PARENT;
 	newnd.flags |= LOOKUP_RENAME_TARGET;
···
 	if (new_dentry == trap)
 		goto exit5;
 
-	error = mnt_want_write(oldnd.path.mnt);
-	if (error)
-		goto exit5;
 	error = security_path_rename(&oldnd.path, old_dentry,
 				     &newnd.path, new_dentry);
 	if (error)
-		goto exit6;
+		goto exit5;
 	error = vfs_rename(old_dir->d_inode, old_dentry,
 				   new_dir->d_inode, new_dentry);
-exit6:
-	mnt_drop_write(oldnd.path.mnt);
 exit5:
 	dput(new_dentry);
 exit4:
 	dput(old_dentry);
 exit3:
 	unlock_rename(new_dir, old_dir);
+	mnt_drop_write(oldnd.path.mnt);
 exit2:
 	path_put(&newnd.path);
 	putname(to);

fs/namespace.c

···
 }
 
 /*
- * Most r/o checks on a fs are for operations that take
- * discrete amounts of time, like a write() or unlink().
- * We must keep track of when those operations start
- * (for permission checks) and when they end, so that
- * we can determine when writes are able to occur to
- * a filesystem.
+ * Most r/o & frozen checks on a fs are for operations that take discrete
+ * amounts of time, like a write() or unlink().  We must keep track of when
+ * those operations start (for permission checks) and when they end, so that we
+ * can determine when writes are able to occur to a filesystem.
  */
 /**
- * mnt_want_write - get write access to a mount
+ * __mnt_want_write - get write access to a mount without freeze protection
  * @m: the mount on which to take a write
  *
- * This tells the low-level filesystem that a write is
- * about to be performed to it, and makes sure that
- * writes are allowed before returning success.  When
- * the write operation is finished, mnt_drop_write()
- * must be called.  This is effectively a refcount.
+ * This tells the low-level filesystem that a write is about to be performed to
+ * it, and makes sure that writes are allowed (mnt it read-write) before
+ * returning success. This operation does not protect against filesystem being
+ * frozen. When the write operation is finished, __mnt_drop_write() must be
+ * called. This is effectively a refcount.
  */
-int mnt_want_write(struct vfsmount *m)
+int __mnt_want_write(struct vfsmount *m)
 {
 	struct mount *mnt = real_mount(m);
 	int ret = 0;
···
 		ret = -EROFS;
 	}
 	preempt_enable();
+
+	return ret;
+}
+
+/**
+ * mnt_want_write - get write access to a mount
+ * @m: the mount on which to take a write
+ *
+ * This tells the low-level filesystem that a write is about to be performed to
+ * it, and makes sure that writes are allowed (mount is read-write, filesystem
+ * is not frozen) before returning success.  When the write operation is
+ * finished, mnt_drop_write() must be called.  This is effectively a refcount.
+ */
+int mnt_want_write(struct vfsmount *m)
+{
+	int ret;
+
+	sb_start_write(m->mnt_sb);
+	ret = __mnt_want_write(m);
+	if (ret)
+		sb_end_write(m->mnt_sb);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mnt_want_write);
···
 EXPORT_SYMBOL_GPL(mnt_clone_write);
 
 /**
+ * __mnt_want_write_file - get write access to a file's mount
+ * @file: the file who's mount on which to take a write
+ *
+ * This is like __mnt_want_write, but it takes a file and can
+ * do some optimisations if the file is open for write already
+ */
+int __mnt_want_write_file(struct file *file)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+
+	if (!(file->f_mode & FMODE_WRITE) || special_file(inode->i_mode))
+		return __mnt_want_write(file->f_path.mnt);
+	else
+		return mnt_clone_write(file->f_path.mnt);
+}
+
+/**
  * mnt_want_write_file - get write access to a file's mount
  * @file: the file who's mount on which to take a write
  *
···
  */
 int mnt_want_write_file(struct file *file)
 {
-	struct inode *inode = file->f_dentry->d_inode;
-	if (!(file->f_mode & FMODE_WRITE) || special_file(inode->i_mode))
-		return mnt_want_write(file->f_path.mnt);
-	else
-		return mnt_clone_write(file->f_path.mnt);
+	int ret;
+
+	sb_start_write(file->f_path.mnt->mnt_sb);
+	ret = __mnt_want_write_file(file);
+	if (ret)
+		sb_end_write(file->f_path.mnt->mnt_sb);
+	return ret;
 }
 EXPORT_SYMBOL_GPL(mnt_want_write_file);
 
 /**
- * mnt_drop_write - give up write access to a mount
+ * __mnt_drop_write - give up write access to a mount
  * @mnt: the mount on which to give up write access
  *
  * Tells the low-level filesystem that we are done
  * performing writes to it.  Must be matched with
- * mnt_want_write() call above.
+ * __mnt_want_write() call above.
  */
-void mnt_drop_write(struct vfsmount *mnt)
+void __mnt_drop_write(struct vfsmount *mnt)
 {
 	preempt_disable();
 	mnt_dec_writers(real_mount(mnt));
 	preempt_enable();
 }
+
+/**
+ * mnt_drop_write - give up write access to a mount
+ * @mnt: the mount on which to give up write access
+ *
+ * Tells the low-level filesystem that we are done performing writes to it and
+ * also allows filesystem to be frozen again.  Must be matched with
+ * mnt_want_write() call above.
+ */
+void mnt_drop_write(struct vfsmount *mnt)
+{
+	__mnt_drop_write(mnt);
+	sb_end_write(mnt->mnt_sb);
+}
 EXPORT_SYMBOL_GPL(mnt_drop_write);
+
+void __mnt_drop_write_file(struct file *file)
+{
+	__mnt_drop_write(file->f_path.mnt);
+}
 
 void mnt_drop_write_file(struct file *file)
 {

fs/nfsd/nfs4recover.c

···
 	if (status < 0)
 		return;
 
+	status = mnt_want_write_file(rec_file);
+	if (status)
+		return;
+
 	dir = rec_file->f_path.dentry;
 	/* lock the parent */
 	mutex_lock(&dir->d_inode->i_mutex);
···
 		 * as well be forgiving and just succeed silently.
 		 */
 		goto out_put;
-	status = mnt_want_write_file(rec_file);
-	if (status)
-		goto out_put;
 	status = vfs_mkdir(dir->d_inode, dentry, S_IRWXU);
-	mnt_drop_write_file(rec_file);
 out_put:
 	dput(dentry);
 out_unlock:
···
 				" (err %d); please check that %s exists"
 				" and is writeable", status,
 				user_recovery_dirname);
+	mnt_drop_write_file(rec_file);
 	nfs4_reset_creds(original_cred);
 }
 

fs/nfsd/nfsfh.c

···
 		fhp->fh_post_saved = 0;
 #endif
 	}
+	fh_drop_write(fhp);
 	if (exp) {
 		exp_put(exp);
 		fhp->fh_export = NULL;

fs/nfsd/nfsproc.c

···
 	struct dentry	*dchild;
 	int		type, mode;
 	__be32		nfserr;
+	int		hosterr;
 	dev_t		rdev = 0, wanted = new_decode_dev(attr->ia_size);
 
 	dprintk("nfsd: CREATE   %s %.*s\n",
···
 	nfserr = nfserr_exist;
 	if (isdotent(argp->name, argp->len))
 		goto done;
+	hosterr = fh_want_write(dirfhp);
+	if (hosterr) {
+		nfserr = nfserrno(hosterr);
+		goto done;
+	}
+
 	fh_lock_nested(dirfhp, I_MUTEX_PARENT);
 	dchild = lookup_one_len(argp->name, dirfhp->fh_dentry, argp->len);
 	if (IS_ERR(dchild)) {
···
 out_unlock:
 	/* We don't really need to unlock, as fh_put does it. */
 	fh_unlock(dirfhp);
-
+	fh_drop_write(dirfhp);
 done:
 	fh_put(dirfhp);
 	return nfsd_return_dirop(nfserr, resp);

fs/nfsd/vfs.c

···
 	 * If it has, the parent directory should already be locked.
 	 */
 	if (!resfhp->fh_dentry) {
+		host_err = fh_want_write(fhp);
+		if (host_err)
+			goto out_nfserr;
+
 		/* called from nfsd_proc_mkdir, or possibly nfsd3_proc_create */
 		fh_lock_nested(fhp, I_MUTEX_PARENT);
 		dchild = lookup_one_len(fname, dentry, flen);
···
 		goto out;
 	}
 
-	host_err = fh_want_write(fhp);
-	if (host_err)
-		goto out_nfserr;
-
 	/*
 	 * Get the dir op function pointer.
 	 */
 	err = 0;
+	host_err = 0;
 	switch (type) {
 	case S_IFREG:
 		host_err = vfs_create(dirp, dchild, iap->ia_mode, true);
···
 		host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev);
 		break;
 	}
-	if (host_err < 0) {
-		fh_drop_write(fhp);
+	if (host_err < 0)
 		goto out_nfserr;
-	}
 
 	err = nfsd_create_setattr(rqstp, resfhp, iap);
 
···
 	err2 = nfserrno(commit_metadata(fhp));
 	if (err2)
 		err = err2;
-	fh_drop_write(fhp);
 	/*
 	 * Update the file handle to get the new inode info.
 	 */
···
 	err = nfserr_notdir;
 	if (!dirp->i_op->lookup)
 		goto out;
+
+	host_err = fh_want_write(fhp);
+	if (host_err)
+		goto out_nfserr;
+
 	fh_lock_nested(fhp, I_MUTEX_PARENT);
 
 	/*
···
 		v_atime = verifier[1]&0x7fffffff;
 	}
 	
-	host_err = fh_want_write(fhp);
-	if (host_err)
-		goto out_nfserr;
 	if (dchild->d_inode) {
 		err = 0;
 
···
 	if (!err)
 		err = nfserrno(commit_metadata(fhp));
 
-	fh_drop_write(fhp);
 	/*
 	 * Update the filehandle to get the new inode info.
 	 */
···
 	fh_unlock(fhp);
 	if (dchild && !IS_ERR(dchild))
 		dput(dchild);
+	fh_drop_write(fhp);
  	return err;
  
  out_nfserr:
···
 	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
 	if (err)
 		goto out;
+
+	host_err = fh_want_write(fhp);
+	if (host_err)
+		goto out_nfserr;
+
 	fh_lock(fhp);
 	dentry = fhp->fh_dentry;
 	dnew = lookup_one_len(fname, dentry, flen);
 	host_err = PTR_ERR(dnew);
 	if (IS_ERR(dnew))
-		goto out_nfserr;
-
-	host_err = fh_want_write(fhp);
-	if (host_err)
 		goto out_nfserr;
 
 	if (unlikely(path[plen] != 0)) {
···
 	if (isdotent(name, len))
 		goto out;
 
+	host_err = fh_want_write(tfhp);
+	if (host_err) {
+		err = nfserrno(host_err);
+		goto out;
+	}
+
 	fh_lock_nested(ffhp, I_MUTEX_PARENT);
 	ddir = ffhp->fh_dentry;
 	dirp = ddir->d_inode;
···
 
 	dold = tfhp->fh_dentry;
 
-	host_err = fh_want_write(tfhp);
-	if (host_err) {
-		err = nfserrno(host_err);
-		goto out_dput;
-	}
 	err = nfserr_noent;
 	if (!dold->d_inode)
-		goto out_drop_write;
+		goto out_dput;
 	host_err = nfsd_break_lease(dold->d_inode);
 	if (host_err) {
 		err = nfserrno(host_err);
-		goto out_drop_write;
+		goto out_dput;
 	}
 	host_err = vfs_link(dold, dirp, dnew);
 	if (!host_err) {
···
 		else
 			err = nfserrno(host_err);
 	}
-out_drop_write:
-	fh_drop_write(tfhp);
 out_dput:
 	dput(dnew);
 out_unlock:
 	fh_unlock(ffhp);
+	fh_drop_write(tfhp);
 out:
 	return err;
 
···
 	if (!flen || isdotent(fname, flen) || !tlen || isdotent(tname, tlen))
 		goto out;
 
+	host_err = fh_want_write(ffhp);
+	if (host_err) {
+		err = nfserrno(host_err);
+		goto out;
+	}
+
 	/* cannot use fh_lock as we need deadlock protective ordering
 	 * so do it by hand */
 	trap = lock_rename(tdentry, fdentry);
···
 	host_err = -EXDEV;
 	if (ffhp->fh_export->ex_path.mnt != tfhp->fh_export->ex_path.mnt)
 		goto out_dput_new;
-	host_err = fh_want_write(ffhp);
-	if (host_err)
-		goto out_dput_new;
 
 	host_err = nfsd_break_lease(odentry->d_inode);
 	if (host_err)
-		goto out_drop_write;
+		goto out_dput_new;
 	if (ndentry->d_inode) {
 		host_err = nfsd_break_lease(ndentry->d_inode);
 		if (host_err)
-			goto out_drop_write;
+			goto out_dput_new;
 	}
 	host_err = vfs_rename(fdir, odentry, tdir, ndentry);
 	if (!host_err) {
···
 		if (!host_err)
 			host_err = commit_metadata(ffhp);
 	}
-out_drop_write:
-	fh_drop_write(ffhp);
  out_dput_new:
 	dput(ndentry);
  out_dput_old:
···
 	fill_post_wcc(tfhp);
 	unlock_rename(tdentry, fdentry);
 	ffhp->fh_locked = tfhp->fh_locked = 0;
+	fh_drop_write(ffhp);
 
 out:
 	return err;
···
 	if (err)
 		goto out;
 
+	host_err = fh_want_write(fhp);
+	if (host_err)
+		goto out_nfserr;
+
 	fh_lock_nested(fhp, I_MUTEX_PARENT);
 	dentry = fhp->fh_dentry;
 	dirp = dentry->d_inode;
···
 	if (!type)
 		type = rdentry->d_inode->i_mode & S_IFMT;
 
-	host_err = fh_want_write(fhp);
-	if (host_err)
-		goto out_put;
-
 	host_err = nfsd_break_lease(rdentry->d_inode);
 	if (host_err)
-		goto out_drop_write;
+		goto out_put;
 	if (type != S_IFDIR)
 		host_err = vfs_unlink(dirp, rdentry);
 	else
 		host_err = vfs_rmdir(dirp, rdentry);
 	if (!host_err)
 		host_err = commit_metadata(fhp);
-out_drop_write:
-	fh_drop_write(fhp);
 out_put:
 	dput(rdentry);
 

fs/nfsd/vfs.h

···
 
 static inline int fh_want_write(struct svc_fh *fh)
 {
-	return mnt_want_write(fh->fh_export->ex_path.mnt);
+	int ret = mnt_want_write(fh->fh_export->ex_path.mnt);
+
+	if (!ret)
+		fh->fh_want_write = 1;
+	return ret;
 }
 
 static inline void fh_drop_write(struct svc_fh *fh)
 {
-	mnt_drop_write(fh->fh_export->ex_path.mnt);
+	if (fh->fh_want_write) {
+		fh->fh_want_write = 0;
+		mnt_drop_write(fh->fh_export->ex_path.mnt);
+	}
 }
 
 #endif /* LINUX_NFSD_VFS_H */

fs/nilfs2/file.c

···
 	struct page *page = vmf->page;
 	struct inode *inode = vma->vm_file->f_dentry->d_inode;
 	struct nilfs_transaction_info ti;
-	int ret;
+	int ret = 0;
 
 	if (unlikely(nilfs_near_disk_full(inode->i_sb->s_fs_info)))
 		return VM_FAULT_SIGBUS; /* -ENOSPC */
 
+	sb_start_pagefault(inode->i_sb);
 	lock_page(page);
 	if (page->mapping != inode->i_mapping ||
 	    page_offset(page) >= i_size_read(inode) || !PageUptodate(page)) {
 		unlock_page(page);
-		return VM_FAULT_NOPAGE; /* make the VM retry the fault */
+		ret = -EFAULT;	/* make the VM retry the fault */
+		goto out;
 	}
 
 	/*
···
 	ret = nilfs_transaction_begin(inode->i_sb, &ti, 1);
 	/* never returns -ENOMEM, but may return -ENOSPC */
 	if (unlikely(ret))
-		return VM_FAULT_SIGBUS;
+		goto out;
 
-	ret = block_page_mkwrite(vma, vmf, nilfs_get_block);
-	if (ret != VM_FAULT_LOCKED) {
+	ret = __block_page_mkwrite(vma, vmf, nilfs_get_block);
+	if (ret) {
 		nilfs_transaction_abort(inode->i_sb);
-		return ret;
+		goto out;
 	}
 	nilfs_set_file_dirty(inode, 1 << (PAGE_SHIFT - inode->i_blkbits));
 	nilfs_transaction_commit(inode->i_sb);
 
  mapped:
 	wait_on_page_writeback(page);
-	return VM_FAULT_LOCKED;
+ out:
+	sb_end_pagefault(inode->i_sb);
+	return block_page_mkwrite_return(ret);
 }
 
 static const struct vm_operations_struct nilfs_file_vm_ops = {

fs/nilfs2/ioctl.c

···
 		goto out_free;
 	}
 
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
-
 	ret = nilfs_ioctl_move_blocks(inode->i_sb, &argv[0], kbufs[0]);
 	if (ret < 0)
 		printk(KERN_ERR "NILFS: GC failed during preparation: "

fs/nilfs2/segment.c

···
 	if (ret > 0)
 		return 0;
 
-	vfs_check_frozen(sb, SB_FREEZE_WRITE);
+	sb_start_intwrite(sb);
 
 	nilfs = sb->s_fs_info;
 	down_read(&nilfs->ns_segctor_sem);
···
 	current->journal_info = ti->ti_save;
 	if (ti->ti_flags & NILFS_TI_DYNAMIC_ALLOC)
 		kmem_cache_free(nilfs_transaction_cachep, ti);
+	sb_end_intwrite(sb);
 	return ret;
 }
 
···
 		err = nilfs_construct_segment(sb);
 	if (ti->ti_flags & NILFS_TI_DYNAMIC_ALLOC)
 		kmem_cache_free(nilfs_transaction_cachep, ti);
+	sb_end_intwrite(sb);
 	return err;
 }
 
···
 	current->journal_info = ti->ti_save;
 	if (ti->ti_flags & NILFS_TI_DYNAMIC_ALLOC)
 		kmem_cache_free(nilfs_transaction_cachep, ti);
+	sb_end_intwrite(sb);
 }
 
 void nilfs_relax_pressure_in_lock(struct super_block *sb)

fs/ntfs/file.c

···
 	if (err)
 		return err;
 	pos = *ppos;
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
 	/* We can write back this queue in page reclaim. */
 	current->backing_dev_info = mapping->backing_dev_info;
 	written = 0;
···
 
 	BUG_ON(iocb->ki_pos != pos);
 
+	sb_start_write(inode->i_sb);
 	mutex_lock(&inode->i_mutex);
 	ret = ntfs_file_aio_write_nolock(iocb, iov, nr_segs, &iocb->ki_pos);
 	mutex_unlock(&inode->i_mutex);
···
 		if (err < 0)
 			ret = err;
 	}
+	sb_end_write(inode->i_sb);
 	return ret;
 }
 

fs/ocfs2/file.c

···
 {
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	int ret;
 
 	if ((cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) &&
 	    !ocfs2_writes_unwritten_extents(osb))
···
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EBADF;
 
-	return __ocfs2_change_file_space(file, inode, file->f_pos, cmd, sr, 0);
+	ret = mnt_want_write_file(file);
+	if (ret)
+		return ret;
+	ret = __ocfs2_change_file_space(file, inode, file->f_pos, cmd, sr, 0);
+	mnt_drop_write_file(file);
+	return ret;
 }
 
 static long ocfs2_fallocate(struct file *file, int mode, loff_t offset,
···
 	if (iocb->ki_left == 0)
 		return 0;
 
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
+	sb_start_write(inode->i_sb);
 
 	appending = file->f_flags & O_APPEND ? 1 : 0;
 	direct_io = file->f_flags & O_DIRECT ? 1 : 0;
···
 		ocfs2_iocb_clear_sem_locked(iocb);
 
 	mutex_unlock(&inode->i_mutex);
+	sb_end_write(inode->i_sb);
 
 	if (written)
 		ret = written;

fs/ocfs2/ioctl.c

···
 		if (get_user(new_clusters, (int __user *)arg))
 			return -EFAULT;
 
-		return ocfs2_group_extend(inode, new_clusters);
+		status = mnt_want_write_file(filp);
+		if (status)
+			return status;
+		status = ocfs2_group_extend(inode, new_clusters);
+		mnt_drop_write_file(filp);
+		return status;
 	case OCFS2_IOC_GROUP_ADD:
 	case OCFS2_IOC_GROUP_ADD64:
 		if (!capable(CAP_SYS_RESOURCE))
···
 		if (copy_from_user(&input, (int __user *) arg, sizeof(input)))
 			return -EFAULT;
 
-		return ocfs2_group_add(inode, &input);
+		status = mnt_want_write_file(filp);
+		if (status)
+			return status;
+		status = ocfs2_group_add(inode, &input);
+		mnt_drop_write_file(filp);
+		return status;
 	case OCFS2_IOC_REFLINK:
 		if (copy_from_user(&args, argp, sizeof(args)))
 			return -EFAULT;

fs/ocfs2/journal.c

···
 	if (journal_current_handle())
 		return jbd2_journal_start(journal, max_buffs);
 
+	sb_start_intwrite(osb->sb);
+
 	down_read(&osb->journal->j_trans_barrier);
 
 	handle = jbd2_journal_start(journal, max_buffs);
 	if (IS_ERR(handle)) {
 		up_read(&osb->journal->j_trans_barrier);
+		sb_end_intwrite(osb->sb);
 
 		mlog_errno(PTR_ERR(handle));
 
···
 	if (ret < 0)
 		mlog_errno(ret);
 
-	if (!nested)
+	if (!nested) {
 		up_read(&journal->j_trans_barrier);
+		sb_end_intwrite(osb->sb);
+	}
 
 	return ret;
 }

fs/ocfs2/mmap.c

···
 	sigset_t oldset;
 	int ret;
 
+	sb_start_pagefault(inode->i_sb);
 	ocfs2_block_signals(&oldset);
 
 	/*
···
 
 out:
 	ocfs2_unblock_signals(&oldset);
+	sb_end_pagefault(inode->i_sb);
 	return ret;
 }
 

fs/ocfs2/refcounttree.c

···
 		goto out_dput;
 	}
 
-	error = mnt_want_write(new_path.mnt);
-	if (error) {
-		mlog_errno(error);
-		goto out_dput;
-	}
-
 	error = ocfs2_vfs_reflink(old_path.dentry,
 				  new_path.dentry->d_inode,
 				  new_dentry, preserve);
-	mnt_drop_write(new_path.mnt);
 out_dput:
-	dput(new_dentry);
-	mutex_unlock(&new_path.dentry->d_inode->i_mutex);
-	path_put(&new_path);
+	done_path_create(&new_path, new_dentry);
 out:
 	path_put(&old_path);
 

fs/open.c

···
 	if (IS_APPEND(inode))
 		goto out_putf;
 
+	sb_start_write(inode->i_sb);
 	error = locks_verify_truncate(inode, file, length);
 	if (!error)
 		error = security_path_truncate(&file->f_path);
 	if (!error)
 		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file);
+	sb_end_write(inode->i_sb);
 out_putf:
 	fput(file);
 out:
···
 	if (!file->f_op->fallocate)
 		return -EOPNOTSUPP;
 
-	return file->f_op->fallocate(file, mode, offset, len);
+	sb_start_write(inode->i_sb);
+	ret = file->f_op->fallocate(file, mode, offset, len);
+	sb_end_write(inode->i_sb);
+	return ret;
 }
 
 SYSCALL_DEFINE(fallocate)(int fd, int mode, loff_t offset, loff_t len)
···
 		/*
 		 * Balanced in __fput()
 		 */
-		error = mnt_want_write(mnt);
+		error = __mnt_want_write(mnt);
 		if (error)
 			put_write_access(inode);
 	}
···
 	if (unlikely(f->f_flags & O_PATH))
 		f->f_mode = FMODE_PATH;
 
+	path_get(&f->f_path);
 	inode = f->f_path.dentry->d_inode;
 	if (f->f_mode & FMODE_WRITE) {
 		error = __get_file_write_access(inode, f->f_path.mnt);
···
 	int error;
 	BUG_ON(*opened & FILE_OPENED); /* once it's opened, it's opened */
 
-	mntget(file->f_path.mnt);
-	file->f_path.dentry = dget(dentry);
-
+	file->f_path.dentry = dentry;
 	error = do_dentry_open(file, open, current_cred());
 	if (!error)
 		*opened |= FILE_OPENED;
···
 
 	f->f_flags = flags;
 	f->f_path = *path;
-	path_get(&f->f_path);
 	error = do_dentry_open(f, NULL, cred);
 	if (!error) {
 		error = open_check_o_direct(f);

fs/pipe.c

···
 	return NULL;
 }
 
-struct file *create_write_pipe(int flags)
+int create_pipe_files(struct file **res, int flags)
 {
 	int err;
-	struct inode *inode;
+	struct inode *inode = get_pipe_inode();
 	struct file *f;
 	struct path path;
-	struct qstr name = { .name = "" };
+	static struct qstr name = { .name = "" };
 
-	err = -ENFILE;
-	inode = get_pipe_inode();
 	if (!inode)
-		goto err;
+		return -ENFILE;
 
 	err = -ENOMEM;
 	path.dentry = d_alloc_pseudo(pipe_mnt->mnt_sb, &name);
···
 	f = alloc_file(&path, FMODE_WRITE, &write_pipefifo_fops);
 	if (!f)
 		goto err_dentry;
-	f->f_mapping = inode->i_mapping;
 
 	f->f_flags = O_WRONLY | (flags & (O_NONBLOCK | O_DIRECT));
-	f->f_version = 0;
 
-	return f;
+	res[0] = alloc_file(&path, FMODE_READ, &read_pipefifo_fops);
+	if (!res[0])
+		goto err_file;
 
- err_dentry:
+	path_get(&path);
+	res[0]->f_flags = O_RDONLY | (flags & O_NONBLOCK);
+	res[1] = f;
+	return 0;
+
+err_file:
+	put_filp(f);
+err_dentry:
 	free_pipe_info(inode);
 	path_put(&path);
-	return ERR_PTR(err);
+	return err;
 
- err_inode:
+err_inode:
 	free_pipe_info(inode);
 	iput(inode);
- err:
-	return ERR_PTR(err);
-}
-
-void free_write_pipe(struct file *f)
-{
-	free_pipe_info(f->f_dentry->d_inode);
-	path_put(&f->f_path);
-	put_filp(f);
-}
-
-struct file *create_read_pipe(struct file *wrf, int flags)
-{
-	/* Grab pipe from the writer */
-	struct file *f = alloc_file(&wrf->f_path, FMODE_READ,
-				    &read_pipefifo_fops);
-	if (!f)
-		return ERR_PTR(-ENFILE);
-
-	path_get(&wrf->f_path);
-	f->f_flags = O_RDONLY | (flags & O_NONBLOCK);
-
-	return f;
+	return err;
 }
 
 int do_pipe_flags(int *fd, int flags)
 {
-	struct file *fw, *fr;
+	struct file *files[2];
 	int error;
 	int fdw, fdr;
 
 	if (flags & ~(O_CLOEXEC | O_NONBLOCK | O_DIRECT))
 		return -EINVAL;
 
-	fw = create_write_pipe(flags);
-	if (IS_ERR(fw))
-		return PTR_ERR(fw);
-	fr = create_read_pipe(fw, flags);
-	error = PTR_ERR(fr);
-	if (IS_ERR(fr))
-		goto err_write_pipe;
+	error = create_pipe_files(files, flags);
+	if (error)
+		return error;
 
 	error = get_unused_fd_flags(flags);
 	if (error < 0)
···
 	fdw = error;
 
 	audit_fd_pair(fdr, fdw);
-	fd_install(fdr, fr);
-	fd_install(fdw, fw);
+	fd_install(fdr, files[0]);
+	fd_install(fdw, files[1]);
 	fd[0] = fdr;
 	fd[1] = fdw;
 
···
  err_fdr:
 	put_unused_fd(fdr);
  err_read_pipe:
-	path_put(&fr->f_path);
-	put_filp(fr);
- err_write_pipe:
-	free_write_pipe(fw);
+	fput(files[0]);
+	fput(files[1]);
 	return error;
 }
 

fs/splice.c

···
 	};
 	ssize_t ret;
 
+	sb_start_write(inode->i_sb);
+
 	pipe_lock(pipe);
 
 	splice_from_pipe_begin(&sd);
···
 			*ppos += ret;
 		balance_dirty_pages_ratelimited_nr(mapping, nr_pages);
 	}
+	sb_end_write(inode->i_sb);
 
 	return ret;
 }

fs/super.c

···
 #include <linux/rculist_bl.h>
 #include <linux/cleancache.h>
 #include <linux/fsnotify.h>
+#include <linux/lockdep.h>
 #include "internal.h"
 
 
 LIST_HEAD(super_blocks);
 DEFINE_SPINLOCK(sb_lock);
+
+static char *sb_writers_name[SB_FREEZE_LEVELS] = {
+	"sb_writers",
+	"sb_pagefaults",
+	"sb_internal",
+};
 
 /*
  * One thing we have to be careful of with a per-sb shrinker is that we don't
···
 	return total_objects;
 }
 
+static int init_sb_writers(struct super_block *s, struct file_system_type *type)
+{
+	int err;
+	int i;
+
+	for (i = 0; i < SB_FREEZE_LEVELS; i++) {
+		err = percpu_counter_init(&s->s_writers.counter[i], 0);
+		if (err < 0)
+			goto err_out;
+		lockdep_init_map(&s->s_writers.lock_map[i], sb_writers_name[i],
+				 &type->s_writers_key[i], 0);
+	}
+	init_waitqueue_head(&s->s_writers.wait);
+	init_waitqueue_head(&s->s_writers.wait_unfrozen);
+	return 0;
+err_out:
+	while (--i >= 0)
+		percpu_counter_destroy(&s->s_writers.counter[i]);
+	return err;
+}
+
+static void destroy_sb_writers(struct super_block *s)
+{
+	int i;
+
+	for (i = 0; i < SB_FREEZE_LEVELS; i++)
+		percpu_counter_destroy(&s->s_writers.counter[i]);
+}
+
 /**
  *	alloc_super	-	create new superblock
  *	@type:	filesystem type superblock should belong to
···
 
 	if (s) {
 		if (security_sb_alloc(s)) {
+			/*
+			 * We cannot call security_sb_free() without
+			 * security_sb_alloc() succeeding. So bail out manually
+			 */
 			kfree(s);
 			s = NULL;
 			goto out;
 		}
 #ifdef CONFIG_SMP
 		s->s_files = alloc_percpu(struct list_head);
-		if (!s->s_files) {
-			security_sb_free(s);
-			kfree(s);
-			s = NULL;
-			goto out;
-		} else {
+		if (!s->s_files)
+			goto err_out;
+		else {
 			int i;
 
 			for_each_possible_cpu(i)
···
 #else
 		INIT_LIST_HEAD(&s->s_files);
 #endif
+		if (init_sb_writers(s, type))
+			goto err_out;
 		s->s_flags = flags;
 		s->s_bdi = &default_backing_dev_info;
 		INIT_HLIST_NODE(&s->s_instances);
···
 		mutex_init(&s->s_dquot.dqio_mutex);
 		mutex_init(&s->s_dquot.dqonoff_mutex);
 		init_rwsem(&s->s_dquot.dqptr_sem);
-		init_waitqueue_head(&s->s_wait_unfrozen);
 		s->s_maxbytes = MAX_NON_LFS;
 		s->s_op = &default_op;
 		s->s_time_gran = 1000000000;
···
 	}
 out:
 	return s;
+err_out:
+	security_sb_free(s);
+#ifdef CONFIG_SMP
+	if (s->s_files)
+		free_percpu(s->s_files);
+#endif
+	destroy_sb_writers(s);
+	kfree(s);
+	s = NULL;
+	goto out;
 }
 
 /**
···
 #ifdef CONFIG_SMP
 	free_percpu(s->s_files);
 #endif
+	destroy_sb_writers(s);
 	security_sb_free(s);
 	WARN_ON(!list_empty(&s->s_mounts));
 	kfree(s->s_subtype);
···
 {
 	while (1) {
 		struct super_block *s = get_super(bdev);
-		if (!s || s->s_frozen == SB_UNFROZEN)
+		if (!s || s->s_writers.frozen == SB_UNFROZEN)
 			return s;
 		up_read(&s->s_umount);
-		vfs_check_frozen(s, SB_FREEZE_WRITE);
+		wait_event(s->s_writers.wait_unfrozen,
+			   s->s_writers.frozen == SB_UNFROZEN);
 		put_super(s);
 	}
 }
···
 	int retval;
 	int remount_ro;
 
-	if (sb->s_frozen != SB_UNFROZEN)
+	if (sb->s_writers.frozen != SB_UNFROZEN)
 		return -EBUSY;
 
 #ifdef CONFIG_BLOCK
···
 	return ERR_PTR(error);
 }
 
+/*
+ * This is an internal function, please use sb_end_{write,pagefault,intwrite}
+ * instead.
+ */
+void __sb_end_write(struct super_block *sb, int level)
+{
+	percpu_counter_dec(&sb->s_writers.counter[level-1]);
+	/*
+	 * Make sure s_writers are updated before we wake up waiters in
+	 * freeze_super().
+	 */
+	smp_mb();
+	if (waitqueue_active(&sb->s_writers.wait))
+		wake_up(&sb->s_writers.wait);
+	rwsem_release(&sb->s_writers.lock_map[level-1], 1, _RET_IP_);
+}
+EXPORT_SYMBOL(__sb_end_write);
+
+#ifdef CONFIG_LOCKDEP
+/*
+ * We want lockdep to tell us about possible deadlocks with freezing but
+ * it's it bit tricky to properly instrument it. Getting a freeze protection
+ * works as getting a read lock but there are subtle problems. XFS for example
+ * gets freeze protection on internal level twice in some cases, which is OK
+ * only because we already hold a freeze protection also on higher level. Due
+ * to these cases we have to tell lockdep we are doing trylock when we
+ * already hold a freeze protection for a higher freeze level.
+ */
+static void acquire_freeze_lock(struct super_block *sb, int level, bool trylock,
+				unsigned long ip)
+{
+	int i;
+
+	if (!trylock) {
+		for (i = 0; i < level - 1; i++)
+			if (lock_is_held(&sb->s_writers.lock_map[i])) {
+				trylock = true;
+				break;
+			}
+	}
+	rwsem_acquire_read(&sb->s_writers.lock_map[level-1], 0, trylock, ip);
+}
+#endif
+
+/*
+ * This is an internal function, please use sb_start_{write,pagefault,intwrite}
+ * instead.
+ */
+int __sb_start_write(struct super_block *sb, int level, bool wait)
+{
+retry:
+	if (unlikely(sb->s_writers.frozen >= level)) {
+		if (!wait)
+			return 0;
+		wait_event(sb->s_writers.wait_unfrozen,
+			   sb->s_writers.frozen < level);
+	}
+
+#ifdef CONFIG_LOCKDEP
+	acquire_freeze_lock(sb, level, !wait, _RET_IP_);
+#endif
+	percpu_counter_inc(&sb->s_writers.counter[level-1]);
+	/*
+	 * Make sure counter is updated before we check for frozen.
+	 * freeze_super() first sets frozen and then checks the counter.
+	 */
+	smp_mb();
+	if (unlikely(sb->s_writers.frozen >= level)) {
+		__sb_end_write(sb, level);
+		goto retry;
+	}
+	return 1;
+}
+EXPORT_SYMBOL(__sb_start_write);
+
+/**
+ * sb_wait_write - wait until all writers to given file system finish
+ * @sb: the super for which we wait
+ * @level: type of writers we wait for (normal vs page fault)
+ *
+ * This function waits until there are no writers of given type to given file
+ * system. Caller of this function should make sure there can be no new writers
+ * of type @level before calling this function. Otherwise this function can
+ * livelock.
+ */
+static void sb_wait_write(struct super_block *sb, int level)
+{
+	s64 writers;
+
+	/*
+	 * We just cycle-through lockdep here so that it does not complain
+	 * about returning with lock to userspace
+	 */
+	rwsem_acquire(&sb->s_writers.lock_map[level-1], 0, 0, _THIS_IP_);
+	rwsem_release(&sb->s_writers.lock_map[level-1], 1, _THIS_IP_);
+
+	do {
+		DEFINE_WAIT(wait);
+
+		/*
+		 * We use a barrier in prepare_to_wait() to separate setting
+		 * of frozen and checking of the counter
+		 */
+		prepare_to_wait(&sb->s_writers.wait, &wait,
+				TASK_UNINTERRUPTIBLE);
+
+		writers = percpu_counter_sum(&sb->s_writers.counter[level-1]);
+		if (writers)
+			schedule();
+
+		finish_wait(&sb->s_writers.wait, &wait);
+	} while (writers);
+}
+
 /**
  * freeze_super - lock the filesystem and force it into a consistent state
  * @sb: the super to lock
···
  * Syncs the super to make sure the filesystem is consistent and calls the fs's
  * freeze_fs.  Subsequent calls to this without first thawing the fs will return
  * -EBUSY.
+ *
+ * During this function, sb->s_writers.frozen goes through these values:
+ *
+ * SB_UNFROZEN: File system is normal, all writes progress as usual.
+ *
+ * SB_FREEZE_WRITE: The file system is in the process of being frozen.  New
+ * writes should be blocked, though page faults are still allowed. We wait for
+ * all writes to complete and then proceed to the next stage.
+ *
+ * SB_FREEZE_PAGEFAULT: Freezing continues. Now also page faults are blocked
+ * but internal fs threads can still modify the filesystem (although they
+ * should not dirty new pages or inodes), writeback can run etc. After waiting
+ * for all running page faults we sync the filesystem which will clean all
+ * dirty pages and inodes (no new dirty pages or inodes can be created when
+ * sync is running).
+ *
+ * SB_FREEZE_FS: The file system is frozen. Now all internal sources of fs
+ * modification are blocked (e.g. XFS preallocation truncation on inode
+ * reclaim). This is usually implemented by blocking new transactions for
+ * filesystems that have them and need this additional guard. After all
+ * internal writers are finished we call ->freeze_fs() to finish filesystem
+ * freezing. Then we transition to SB_FREEZE_COMPLETE state. This state is
+ * mostly auxiliary for filesystems to verify they do not modify frozen fs.
+ *
+ * sb->s_writers.frozen is protected by sb->s_umount.
  */
 int freeze_super(struct super_block *sb)
 {
···
 
 	atomic_inc(&sb->s_active);
 	down_write(&sb->s_umount);
-	if (sb->s_frozen) {
+	if (sb->s_writers.frozen != SB_UNFROZEN) {
 		deactivate_locked_super(sb);
 		return -EBUSY;
 	}
···
 	}
 
 	if (sb->s_flags & MS_RDONLY) {
-		sb->s_frozen = SB_FREEZE_TRANS;
-		smp_wmb();
+		/* Nothing to do really... */
+		sb->s_writers.frozen = SB_FREEZE_COMPLETE;
 		up_write(&sb->s_umount);
 		return 0;
 	}
 
-	sb->s_frozen = SB_FREEZE_WRITE;
+	/* From now on, no new normal writers can start */
+	sb->s_writers.frozen = SB_FREEZE_WRITE;
 	smp_wmb();
 
+	/* Release s_umount to preserve sb_start_write -> s_umount ordering */
+	up_write(&sb->s_umount);
+
+	sb_wait_write(sb, SB_FREEZE_WRITE);
+
+	/* Now we go and block page faults... */
+	down_write(&sb->s_umount);
+	sb->s_writers.frozen = SB_FREEZE_PAGEFAULT;
+	smp_wmb();
+
+	sb_wait_write(sb, SB_FREEZE_PAGEFAULT);
+
+	/* All writers are done so after syncing there won't be dirty data */
 	sync_filesystem(sb);
 
-	sb->s_frozen = SB_FREEZE_TRANS;
+	/* Now wait for internal filesystem counter */
+	sb->s_writers.frozen = SB_FREEZE_FS;
 	smp_wmb();
+	sb_wait_write(sb, SB_FREEZE_FS);
 
-	sync_blockdev(sb->s_bdev);
 	if (sb->s_op->freeze_fs) {
 		ret = sb->s_op->freeze_fs(sb);
 		if (ret) {
 			printk(KERN_ERR
 				"VFS:Filesystem freeze failed\n");
-			sb->s_frozen = SB_UNFROZEN;
+			sb->s_writers.frozen = SB_UNFROZEN;
 			smp_wmb();
-			wake_up(&sb->s_wait_unfrozen);
+			wake_up(&sb->s_writers.wait_unfrozen);
 			deactivate_locked_super(sb);
 			return ret;
 		}
 	}
+	/*
+	 * This is just for debugging purposes so that fs can warn if it
+	 * sees write activity when frozen is set to SB_FREEZE_COMPLETE.
+	 */
+	sb->s_writers.frozen = SB_FREEZE_COMPLETE;
 	up_write(&sb->s_umount);
 	return 0;
 }
···
 	int error;
 
 	down_write(&sb->s_umount);
-	if (sb->s_frozen == SB_UNFROZEN) {
+	if (sb->s_writers.frozen == SB_UNFROZEN) {
 		up_write(&sb->s_umount);
 		return -EINVAL;
 	}
···
 		if (error) {
 			printk(KERN_ERR
 				"VFS:Filesystem thaw failed\n");
-			sb->s_frozen = SB_FREEZE_TRANS;
 			up_write(&sb->s_umount);
 			return error;
 		}
 	}
 
 out:
-	sb->s_frozen = SB_UNFROZEN;
+	sb->s_writers.frozen = SB_UNFROZEN;
 	smp_wmb();
-	wake_up(&sb->s_wait_unfrozen);
+	wake_up(&sb->s_writers.wait_unfrozen);
 	deactivate_locked_super(sb);
 
 	return 0;

fs/sysfs/bin.c

···
 	ret = 0;
 	if (bb->vm_ops->page_mkwrite)
 		ret = bb->vm_ops->page_mkwrite(vma, vmf);
+	else
+		file_update_time(file);
 
 	sysfs_put_active(attr_sd);
 	return ret;

fs/xfs/xfs_aops.c

···
 	ioend->io_append_trans = tp;
 
 	/*
+	 * We will pass freeze protection with a transaction.  So tell lockdep
+	 * we released it.
+	 */
+	rwsem_release(&ioend->io_inode->i_sb->s_writers.lock_map[SB_FREEZE_FS-1],
+		      1, _THIS_IP_);
+	/*
 	 * We hand off the transaction to the completion thread now, so
 	 * clear the flag here.
 	 */
···
 	struct xfs_inode *ip = XFS_I(ioend->io_inode);
 	int		error = 0;
 
+	if (ioend->io_append_trans) {
+		/*
+		 * We've got freeze protection passed with the transaction.
+		 * Tell lockdep about it.
+		 */
+		rwsem_acquire_read(
+			&ioend->io_inode->i_sb->s_writers.lock_map[SB_FREEZE_FS-1],
+			0, 1, _THIS_IP_);
+	}
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
 		ioend->io_error = -EIO;
 		goto done;
···
 	if (ioend->io_append_trans) {
 		current_set_flags_nested(&ioend->io_append_trans->t_pflags,
 					 PF_FSTRANS);
+		rwsem_acquire_read(
+			&inode->i_sb->s_writers.lock_map[SB_FREEZE_FS-1],
+			0, 1, _THIS_IP_);
 		xfs_trans_cancel(ioend->io_append_trans, 0);
 	}
 out_destroy_ioend:

fs/xfs/xfs_file.c

···
 	if (ocount == 0)
 		return 0;
 
-	xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE);
+	sb_start_write(inode->i_sb);
 
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return -EIO;
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+		ret = -EIO;
+		goto out;
+	}
 
 	if (unlikely(file->f_flags & O_DIRECT))
 		ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, ocount);
···
 			ret = err;
 	}
 
+out:
+	sb_end_write(inode->i_sb);
 	return ret;
 }
 

fs/xfs/xfs_ioctl.c

···
 	if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t)))
 		return -XFS_ERROR(EFAULT);
 
+	error = mnt_want_write_file(parfilp);
+	if (error)
+		return error;
+
 	dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq);
-	if (IS_ERR(dentry))
+	if (IS_ERR(dentry)) {
+		mnt_drop_write_file(parfilp);
 		return PTR_ERR(dentry);
+	}
 
 	if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) {
 		error = -XFS_ERROR(EPERM);
···
 				 fsd.fsd_dmstate);
 
  out:
+	mnt_drop_write_file(parfilp);
 	dput(dentry);
 	return error;
 }
···
 	if (ioflags & IO_INVIS)
 		attr_flags |= XFS_ATTR_DMI;
 
+	error = mnt_want_write_file(filp);
+	if (error)
+		return error;
 	error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags);
+	mnt_drop_write_file(filp);
 	return -error;
 }
 
···
 {
 	struct fsxattr		fa;
 	unsigned int		mask;
+	int error;
 
 	if (copy_from_user(&fa, arg, sizeof(fa)))
 		return -EFAULT;
···
 	if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
 		mask |= FSX_NONBLOCK;
 
-	return -xfs_ioctl_setattr(ip, &fa, mask);
+	error = mnt_want_write_file(filp);
+	if (error)
+		return error;
+	error = xfs_ioctl_setattr(ip, &fa, mask);
+	mnt_drop_write_file(filp);
+	return -error;
 }
 
 STATIC int
···
 	struct fsxattr		fa;
 	unsigned int		flags;
 	unsigned int		mask;
+	int error;
 
 	if (copy_from_user(&flags, arg, sizeof(flags)))
 		return -EFAULT;
···
 		mask |= FSX_NONBLOCK;
 	fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip));
 
-	return -xfs_ioctl_setattr(ip, &fa, mask);
+	error = mnt_want_write_file(filp);
+	if (error)
+		return error;
+	error = xfs_ioctl_setattr(ip, &fa, mask);
+	mnt_drop_write_file(filp);
+	return -error;
 }
 
 STATIC int
···
 		if (copy_from_user(&dmi, arg, sizeof(dmi)))
 			return -XFS_ERROR(EFAULT);
 
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
+
 		error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
 				dmi.fsd_dmstate);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 
···
 
 		if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t)))
 			return -XFS_ERROR(EFAULT);
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_swapext(&sxp);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 
···
 		if (copy_from_user(&inout, arg, sizeof(inout)))
 			return -XFS_ERROR(EFAULT);
 
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
+
 		/* input parameter is passed in resblks field of structure */
 		in = inout.resblks;
 		error = xfs_reserve_blocks(mp, &in, &inout);
+		mnt_drop_write_file(filp);
 		if (error)
 			return -error;
 
···
 		if (copy_from_user(&in, arg, sizeof(in)))
 			return -XFS_ERROR(EFAULT);
 
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_growfs_data(mp, &in);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 
···
 		if (copy_from_user(&in, arg, sizeof(in)))
 			return -XFS_ERROR(EFAULT);
 
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_growfs_log(mp, &in);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 
···
 		if (copy_from_user(&in, arg, sizeof(in)))
 			return -XFS_ERROR(EFAULT);
 
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_growfs_rt(mp, &in);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 

fs/xfs/xfs_ioctl32.c

···
 
 		if (xfs_compat_growfs_data_copyin(&in, arg))
 			return -XFS_ERROR(EFAULT);
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_growfs_data(mp, &in);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 	case XFS_IOC_FSGROWFSRT_32: {
···
 
 		if (xfs_compat_growfs_rt_copyin(&in, arg))
 			return -XFS_ERROR(EFAULT);
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_growfs_rt(mp, &in);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 #endif
···
 				   offsetof(struct xfs_swapext, sx_stat)) ||
 		    xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat))
 			return -XFS_ERROR(EFAULT);
+		error = mnt_want_write_file(filp);
+		if (error)
+			return error;
 		error = xfs_swapext(&sxp);
+		mnt_drop_write_file(filp);
 		return -error;
 	}
 	case XFS_IOC_FSBULKSTAT_32:

fs/xfs/xfs_iomap.c

···
 		 * the same inode that we complete here and might deadlock
 		 * on the iolock.
 		 */
-		xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
+		sb_start_intwrite(mp->m_super);
 		tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS);
-		tp->t_flags |= XFS_TRANS_RESERVE;
+		tp->t_flags |= XFS_TRANS_RESERVE | XFS_TRANS_FREEZE_PROT;
 		error = xfs_trans_reserve(tp, resblks,
 				XFS_WRITE_LOG_RES(mp), 0,
 				XFS_TRANS_PERM_LOG_RES,

fs/xfs/xfs_mount.c

···
 int
 xfs_fs_writable(xfs_mount_t *mp)
 {
-	return !(xfs_test_for_freeze(mp) || XFS_FORCED_SHUTDOWN(mp) ||
+	return !(mp->m_super->s_writers.frozen || XFS_FORCED_SHUTDOWN(mp) ||
 		(mp->m_flags & XFS_MOUNT_RDONLY));
 }
 

fs/xfs/xfs_mount.h

···
 #define SHUTDOWN_REMOTE_REQ	0x0010	/* shutdown came from remote cell */
 #define SHUTDOWN_DEVICE_REQ	0x0020	/* failed all paths to the device */
 
-#define xfs_test_for_freeze(mp)		((mp)->m_super->s_frozen)
-#define xfs_wait_for_freeze(mp,l)	vfs_check_frozen((mp)->m_super, (l))
-
 /*
  * Flags for xfs_mountfs
  */

fs/xfs/xfs_sync.c

···
 	if (!(mp->m_super->s_flags & MS_ACTIVE) &&
 	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
 		/* dgc: errors ignored here */
-		if (mp->m_super->s_frozen == SB_UNFROZEN &&
+		if (mp->m_super->s_writers.frozen == SB_UNFROZEN &&
 		    xfs_log_need_covered(mp))
 			error = xfs_fs_log_dummy(mp);
 		else

fs/xfs/xfs_trans.c

···
 	xfs_mount_t	*mp,
 	uint		type)
 {
-	xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
-	return _xfs_trans_alloc(mp, type, KM_SLEEP);
+	xfs_trans_t     *tp;
+
+	sb_start_intwrite(mp->m_super);
+	tp = _xfs_trans_alloc(mp, type, KM_SLEEP);
+	tp->t_flags |= XFS_TRANS_FREEZE_PROT;
+	return tp;
 }
 
 xfs_trans_t *
···
 {
 	xfs_trans_t	*tp;
 
+	WARN_ON(mp->m_super->s_writers.frozen == SB_FREEZE_COMPLETE);
 	atomic_inc(&mp->m_active_trans);
 
 	tp = kmem_zone_zalloc(xfs_trans_zone, memflags);
···
 	xfs_extent_busy_clear(tp->t_mountp, &tp->t_busy, false);
 
 	atomic_dec(&tp->t_mountp->m_active_trans);
+	if (tp->t_flags & XFS_TRANS_FREEZE_PROT)
+		sb_end_intwrite(tp->t_mountp->m_super);
 	xfs_trans_free_dqinfo(tp);
 	kmem_zone_free(xfs_trans_zone, tp);
 }
···
 	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
 	ASSERT(tp->t_ticket != NULL);
 
-	ntp->t_flags = XFS_TRANS_PERM_LOG_RES | (tp->t_flags & XFS_TRANS_RESERVE);
+	ntp->t_flags = XFS_TRANS_PERM_LOG_RES |
+		       (tp->t_flags & XFS_TRANS_RESERVE) |
+		       (tp->t_flags & XFS_TRANS_FREEZE_PROT);
+	/* We gave our writer reference to the new transaction */
+	tp->t_flags &= ~XFS_TRANS_FREEZE_PROT;
 	ntp->t_ticket = xfs_log_ticket_get(tp->t_ticket);
 	ntp->t_blk_res = tp->t_blk_res - tp->t_blk_res_used;
 	tp->t_blk_res = tp->t_blk_res_used;

fs/xfs/xfs_trans.h

···
 #define	XFS_TRANS_SYNC		0x08	/* make commit synchronous */
 #define XFS_TRANS_DQ_DIRTY	0x10	/* at least one dquot in trx dirty */
 #define XFS_TRANS_RESERVE	0x20    /* OK to use reserved data blocks */
+#define XFS_TRANS_FREEZE_PROT	0x40	/* Transaction has elevated writer
+					   count in superblock */
 
 /*
  * Values for call flags parameter.

include/linux/audit.h

···
 #define AUDIT_LAST_KERN_ANOM_MSG    1799
 #define AUDIT_ANOM_PROMISCUOUS      1700 /* Device changed promiscuous mode */
 #define AUDIT_ANOM_ABEND            1701 /* Process ended abnormally */
+#define AUDIT_ANOM_LINK		    1702 /* Suspicious use of file links */
 #define AUDIT_INTEGRITY_DATA	    1800 /* Data integrity verification */
 #define AUDIT_INTEGRITY_METADATA    1801 /* Metadata integrity verification */
 #define AUDIT_INTEGRITY_STATUS	    1802 /* Integrity enable status */
···
 					     const struct path *path);
 extern void		    audit_log_key(struct audit_buffer *ab,
 					  char *key);
+extern void		    audit_log_link_denied(const char *operation,
+						  struct path *link);
 extern void		    audit_log_lost(const char *message);
 #ifdef CONFIG_SECURITY
 extern void 		    audit_log_secctx(struct audit_buffer *ab, u32 secid);
···
 #define audit_log_untrustedstring(a,s) do { ; } while (0)
 #define audit_log_d_path(b, p, d) do { ; } while (0)
 #define audit_log_key(b, k) do { ; } while (0)
+#define audit_log_link_denied(o, l) do { ; } while (0)
 #define audit_log_secctx(b,s) do { ; } while (0)
 #define audit_enabled 0
 #endif

include/linux/fs.h

···
 #include <linux/shrinker.h>
 #include <linux/migrate_mode.h>
 #include <linux/uidgid.h>
+#include <linux/lockdep.h>
 
 #include <asm/byteorder.h>
 
···
 extern int sysctl_nr_open;
 extern struct inodes_stat_t inodes_stat;
 extern int leases_enable, lease_break_time;
+extern int sysctl_protected_symlinks;
+extern int sysctl_protected_hardlinks;
 
 struct buffer_head;
 typedef int (get_block_t)(struct inode *inode, sector_t iblock,
···
 extern pid_t f_getown(struct file *filp);
 extern int send_sigurg(struct fown_struct *fown);
 
+struct mm_struct;
+
 /*
  *	Umount options
  */
···
 
 extern struct list_head super_blocks;
 extern spinlock_t sb_lock;
+
+/* Possible states of 'frozen' field */
+enum {
+	SB_UNFROZEN = 0,		/* FS is unfrozen */
+	SB_FREEZE_WRITE	= 1,		/* Writes, dir ops, ioctls frozen */
+	SB_FREEZE_PAGEFAULT = 2,	/* Page faults stopped as well */
+	SB_FREEZE_FS = 3,		/* For internal FS use (e.g. to stop
+					 * internal threads if needed) */
+	SB_FREEZE_COMPLETE = 4,		/* ->freeze_fs finished successfully */
+};
+
+#define SB_FREEZE_LEVELS (SB_FREEZE_COMPLETE - 1)
+
+struct sb_writers {
+	/* Counters for counting writers at each level */
+	struct percpu_counter	counter[SB_FREEZE_LEVELS];
+	wait_queue_head_t	wait;		/* queue for waiting for
+						   writers / faults to finish */
+	int			frozen;		/* Is sb frozen? */
+	wait_queue_head_t	wait_unfrozen;	/* queue for waiting for
+						   sb to be thawed */
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map	lock_map[SB_FREEZE_LEVELS];
+#endif
+};
 
 struct super_block {
 	struct list_head	s_list;		/* Keep this first */
···
 	struct hlist_node	s_instances;
 	struct quota_info	s_dquot;	/* Diskquota specific options */
 
-	int			s_frozen;
-	wait_queue_head_t	s_wait_unfrozen;
+	struct sb_writers	s_writers;
 
 	char s_id[32];				/* Informational name */
 	u8 s_uuid[16];				/* UUID */
···
 /*
  * Snapshotting support.
  */
-enum {
-	SB_UNFROZEN = 0,
-	SB_FREEZE_WRITE	= 1,
-	SB_FREEZE_TRANS = 2,
-};
 
-#define vfs_check_frozen(sb, level) \
-	wait_event((sb)->s_wait_unfrozen, ((sb)->s_frozen < (level)))
+void __sb_end_write(struct super_block *sb, int level);
+int __sb_start_write(struct super_block *sb, int level, bool wait);
+
+/**
+ * sb_end_write - drop write access to a superblock
+ * @sb: the super we wrote to
+ *
+ * Decrement number of writers to the filesystem. Wake up possible waiters
+ * wanting to freeze the filesystem.
+ */
+static inline void sb_end_write(struct super_block *sb)
+{
+	__sb_end_write(sb, SB_FREEZE_WRITE);
+}
+
+/**
+ * sb_end_pagefault - drop write access to a superblock from a page fault
+ * @sb: the super we wrote to
+ *
+ * Decrement number of processes handling write page fault to the filesystem.
+ * Wake up possible waiters wanting to freeze the filesystem.
+ */
+static inline void sb_end_pagefault(struct super_block *sb)
+{
+	__sb_end_write(sb, SB_FREEZE_PAGEFAULT);
+}
+
+/**
+ * sb_end_intwrite - drop write access to a superblock for internal fs purposes
+ * @sb: the super we wrote to
+ *
+ * Decrement fs-internal number of writers to the filesystem.  Wake up possible
+ * waiters wanting to freeze the filesystem.
+ */
+static inline void sb_end_intwrite(struct super_block *sb)
+{
+	__sb_end_write(sb, SB_FREEZE_FS);
+}
+
+/**
+ * sb_start_write - get write access to a superblock
+ * @sb: the super we write to
+ *
+ * When a process wants to write data or metadata to a file system (i.e. dirty
+ * a page or an inode), it should embed the operation in a sb_start_write() -
+ * sb_end_write() pair to get exclusion against file system freezing. This
+ * function increments number of writers preventing freezing. If the file
+ * system is already frozen, the function waits until the file system is
+ * thawed.
+ *
+ * Since freeze protection behaves as a lock, users have to preserve
+ * ordering of freeze protection and other filesystem locks. Generally,
+ * freeze protection should be the outermost lock. In particular, we have:
+ *
+ * sb_start_write
+ *   -> i_mutex			(write path, truncate, directory ops, ...)
+ *   -> s_umount		(freeze_super, thaw_super)
+ */
+static inline void sb_start_write(struct super_block *sb)
+{
+	__sb_start_write(sb, SB_FREEZE_WRITE, true);
+}
+
+static inline int sb_start_write_trylock(struct super_block *sb)
+{
+	return __sb_start_write(sb, SB_FREEZE_WRITE, false);
+}
+
+/**
+ * sb_start_pagefault - get write access to a superblock from a page fault
+ * @sb: the super we write to
+ *
+ * When a process starts handling write page fault, it should embed the
+ * operation into sb_start_pagefault() - sb_end_pagefault() pair to get
+ * exclusion against file system freezing. This is needed since the page fault
+ * is going to dirty a page. This function increments number of running page
+ * faults preventing freezing. If the file system is already frozen, the
+ * function waits until the file system is thawed.
+ *
+ * Since page fault freeze protection behaves as a lock, users have to preserve
+ * ordering of freeze protection and other filesystem locks. It is advised to
+ * put sb_start_pagefault() close to mmap_sem in lock ordering. Page fault
+ * handling code implies lock dependency:
+ *
+ * mmap_sem
+ *   -> sb_start_pagefault
+ */
+static inline void sb_start_pagefault(struct super_block *sb)
+{
+	__sb_start_write(sb, SB_FREEZE_PAGEFAULT, true);
+}
+
+/*
+ * sb_start_intwrite - get write access to a superblock for internal fs purposes
+ * @sb: the super we write to
+ *
+ * This is the third level of protection against filesystem freezing. It is
+ * free for use by a filesystem. The only requirement is that it must rank
+ * below sb_start_pagefault.
+ *
+ * For example filesystem can call sb_start_intwrite() when starting a
+ * transaction which somewhat eases handling of freezing for internal sources
+ * of filesystem changes (internal fs threads, discarding preallocation on file
+ * close, etc.).
+ */
+static inline void sb_start_intwrite(struct super_block *sb)
+{
+	__sb_start_write(sb, SB_FREEZE_FS, true);
+}
+
 
 extern bool inode_owner_or_capable(const struct inode *inode);
 
···
 	struct lock_class_key s_lock_key;
 	struct lock_class_key s_umount_key;
 	struct lock_class_key s_vfs_rename_key;
+	struct lock_class_key s_writers_key[SB_FREEZE_LEVELS];
 
 	struct lock_class_key i_lock_key;
 	struct lock_class_key i_mutex_key;
···
 }
 #endif
 extern int do_pipe_flags(int *, int);
-extern struct file *create_read_pipe(struct file *f, int flags);
-extern struct file *create_write_pipe(int flags);
-extern void free_write_pipe(struct file *);
 
 extern int kernel_read(struct file *, loff_t, char *, unsigned long);
 extern struct file * open_exec(const char *);

include/linux/mm.h

···
 
 /* generic vm_area_ops exported for stackable file systems */
 extern int filemap_fault(struct vm_area_struct *, struct vm_fault *);
+extern int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
 
 /* mm/page-writeback.c */
 int write_one_page(struct page *page, int wait);

include/linux/namei.h

···
 
 extern struct dentry *kern_path_create(int, const char *, struct path *, int);
 extern struct dentry *user_path_create(int, const char __user *, struct path *, int);
+extern void done_path_create(struct path *, struct dentry *);
 extern struct dentry *kern_path_locked(const char *, struct path *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct path *);

include/linux/nfsd/nfsfh.h

···
 	int			fh_maxsize;	/* max size for fh_handle */
 
 	unsigned char		fh_locked;	/* inode locked by us */
+	unsigned char		fh_want_write;	/* remount protection taken */
 
 #ifdef CONFIG_NFSD_V3
 	unsigned char		fh_post_saved;	/* post-op attrs saved */

include/linux/pipe_fs_i.h

···
 long pipe_fcntl(struct file *, unsigned int, unsigned long arg);
 struct pipe_inode_info *get_pipe_info(struct file *file);
 
+int create_pipe_files(struct file **, int);
+
 #endif

kernel/audit.c

···
 }
 
 /**
+ * audit_log_link_denied - report a link restriction denial
+ * @operation: specific link opreation
+ * @link: the path that triggered the restriction
+ */
+void audit_log_link_denied(const char *operation, struct path *link)
+{
+	struct audit_buffer *ab;
+
+	ab = audit_log_start(current->audit_context, GFP_KERNEL,
+			     AUDIT_ANOM_LINK);
+	audit_log_format(ab, "op=%s action=denied", operation);
+	audit_log_format(ab, " pid=%d comm=", current->pid);
+	audit_log_untrustedstring(ab, current->comm);
+	audit_log_d_path(ab, " path=", link);
+	audit_log_format(ab, " dev=");
+	audit_log_untrustedstring(ab, link->dentry->d_inode->i_sb->s_id);
+	audit_log_format(ab, " ino=%lu", link->dentry->d_inode->i_ino);
+	audit_log_end(ab);
+}
+
+/**
  * audit_log_end - end one audit record
  * @ab: the audit_buffer
  *

kernel/sysctl.c

···
 #endif
 #endif
 	{
+		.procname	= "protected_symlinks",
+		.data		= &sysctl_protected_symlinks,
+		.maxlen		= sizeof(int),
+		.mode		= 0600,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
+	{
+		.procname	= "protected_hardlinks",
+		.data		= &sysctl_protected_hardlinks,
+		.maxlen		= sizeof(int),
+		.mode		= 0600,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
+	{
 		.procname	= "suid_dumpable",
 		.data		= &suid_dumpable,
 		.maxlen		= sizeof(int),

lib/percpu_counter.c

···
 
 #ifdef CONFIG_HOTPLUG_CPU
 static LIST_HEAD(percpu_counters);
-static DEFINE_MUTEX(percpu_counters_lock);
+static DEFINE_SPINLOCK(percpu_counters_lock);
 #endif
 
 #ifdef CONFIG_DEBUG_OBJECTS_PERCPU_COUNTER
···
 
 #ifdef CONFIG_HOTPLUG_CPU
 	INIT_LIST_HEAD(&fbc->list);
-	mutex_lock(&percpu_counters_lock);
+	spin_lock(&percpu_counters_lock);
 	list_add(&fbc->list, &percpu_counters);
-	mutex_unlock(&percpu_counters_lock);
+	spin_unlock(&percpu_counters_lock);
 #endif
 	return 0;
 }
···
 	debug_percpu_counter_deactivate(fbc);
 
 #ifdef CONFIG_HOTPLUG_CPU
-	mutex_lock(&percpu_counters_lock);
+	spin_lock(&percpu_counters_lock);
 	list_del(&fbc->list);
-	mutex_unlock(&percpu_counters_lock);
+	spin_unlock(&percpu_counters_lock);
 #endif
 	free_percpu(fbc->counters);
 	fbc->counters = NULL;
···
 		return NOTIFY_OK;
 
 	cpu = (unsigned long)hcpu;
-	mutex_lock(&percpu_counters_lock);
+	spin_lock(&percpu_counters_lock);
 	list_for_each_entry(fbc, &percpu_counters, list) {
 		s32 *pcount;
 		unsigned long flags;
···
 		*pcount = 0;
 		raw_spin_unlock_irqrestore(&fbc->lock, flags);
 	}
-	mutex_unlock(&percpu_counters_lock);
+	spin_unlock(&percpu_counters_lock);
 #endif
 	return NOTIFY_OK;
 }

mm/filemap.c

···
 }
 EXPORT_SYMBOL(filemap_fault);
 
+int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	struct page *page = vmf->page;
+	struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
+	int ret = VM_FAULT_LOCKED;
+
+	sb_start_pagefault(inode->i_sb);
+	file_update_time(vma->vm_file);
+	lock_page(page);
+	if (page->mapping != inode->i_mapping) {
+		unlock_page(page);
+		ret = VM_FAULT_NOPAGE;
+		goto out;
+	}
+	/*
+	 * We mark the page dirty already here so that when freeze is in
+	 * progress, we are guaranteed that writeback during freezing will
+	 * see the dirty page and writeprotect it again.
+	 */
+	set_page_dirty(page);
+out:
+	sb_end_pagefault(inode->i_sb);
+	return ret;
+}
+EXPORT_SYMBOL(filemap_page_mkwrite);
+
 const struct vm_operations_struct generic_file_vm_ops = {
 	.fault		= filemap_fault,
+	.page_mkwrite	= filemap_page_mkwrite,
 };
 
 /* This is used for a general mmap of a disk file */
···
 	count = ocount;
 	pos = *ppos;
 
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
-
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = mapping->backing_dev_info;
 	written = 0;
···
 
 	BUG_ON(iocb->ki_pos != pos);
 
+	sb_start_write(inode->i_sb);
 	mutex_lock(&inode->i_mutex);
 	blk_start_plug(&plug);
 	ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos);
···
 			ret = err;
 	}
 	blk_finish_plug(&plug);
+	sb_end_write(inode->i_sb);
 	return ret;
 }
 EXPORT_SYMBOL(generic_file_aio_write);

mm/filemap_xip.c

···
 
 static const struct vm_operations_struct xip_file_vm_ops = {
 	.fault	= xip_file_fault,
+	.page_mkwrite	= filemap_page_mkwrite,
 };
 
 int xip_file_mmap(struct file * file, struct vm_area_struct * vma)
···
 	loff_t pos;
 	ssize_t ret;
 
+	sb_start_write(inode->i_sb);
+
 	mutex_lock(&inode->i_mutex);
 
 	if (!access_ok(VERIFY_READ, buf, len)) {
···
 
 	pos = *ppos;
 	count = len;
-
-	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
 
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = mapping->backing_dev_info;
···
 	current->backing_dev_info = NULL;
  out_up:
 	mutex_unlock(&inode->i_mutex);
+	sb_end_write(inode->i_sb);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(xip_file_write);

mm/memory.c

···
 		if (!page_mkwrite) {
 			wait_on_page_locked(dirty_page);
 			set_page_dirty_balance(dirty_page, page_mkwrite);
+			/* file_update_time outside page_lock */
+			if (vma->vm_file)
+				file_update_time(vma->vm_file);
 		}
 		put_page(dirty_page);
 		if (page_mkwrite) {
···
 				balance_dirty_pages_ratelimited(mapping);
 			}
 		}
-
-		/* file_update_time outside page_lock */
-		if (vma->vm_file)
-			file_update_time(vma->vm_file);
 
 		return ret;
 	}
···
 
 	if (dirty_page) {
 		struct address_space *mapping = page->mapping;
+		int dirtied = 0;
 
 		if (set_page_dirty(dirty_page))
-			page_mkwrite = 1;
+			dirtied = 1;
 		unlock_page(dirty_page);
 		put_page(dirty_page);
-		if (page_mkwrite && mapping) {
+		if ((dirtied || page_mkwrite) && mapping) {
 			/*
 			 * Some device drivers do not set page.mapping but still
 			 * dirty their pages
···
 		}
 
 		/* file_update_time outside page_lock */
-		if (vma->vm_file)
+		if (vma->vm_file && !page_mkwrite)
 			file_update_time(vma->vm_file);
 	} else {
 		unlock_page(vmf.page);

net/unix/af_unix.c

···
 	return NULL;
 }
 
+static int unix_mknod(const char *sun_path, umode_t mode, struct path *res)
+{
+	struct dentry *dentry;
+	struct path path;
+	int err = 0;
+	/*
+	 * Get the parent directory, calculate the hash for last
+	 * component.
+	 */
+	dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0);
+	err = PTR_ERR(dentry);
+	if (IS_ERR(dentry))
+		return err;
+
+	/*
+	 * All right, let's create it.
+	 */
+	err = security_path_mknod(&path, dentry, mode, 0);
+	if (!err) {
+		err = vfs_mknod(path.dentry->d_inode, dentry, mode, 0);
+		if (!err) {
+			res->mnt = mntget(path.mnt);
+			res->dentry = dget(dentry);
+		}
+	}
+	done_path_create(&path, dentry);
+	return err;
+}
 
 static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
 {
···
 	struct unix_sock *u = unix_sk(sk);
 	struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr;
 	char *sun_path = sunaddr->sun_path;
-	struct dentry *dentry = NULL;
-	struct path path;
 	int err;
 	unsigned int hash;
 	struct unix_address *addr;
···
 	atomic_set(&addr->refcnt, 1);
 
 	if (sun_path[0]) {
-		umode_t mode;
-		err = 0;
-		/*
-		 * Get the parent directory, calculate the hash for last
-		 * component.
-		 */
-		dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0);
-		err = PTR_ERR(dentry);
-		if (IS_ERR(dentry))
-			goto out_mknod_parent;
-
-		/*
-		 * All right, let's create it.
-		 */
-		mode = S_IFSOCK |
+		struct path path;
+		umode_t mode = S_IFSOCK |
 		       (SOCK_INODE(sock)->i_mode & ~current_umask());
-		err = mnt_want_write(path.mnt);
-		if (err)
-			goto out_mknod_dput;
-		err = security_path_mknod(&path, dentry, mode, 0);
-		if (err)
-			goto out_mknod_drop_write;
-		err = vfs_mknod(path.dentry->d_inode, dentry, mode, 0);
-out_mknod_drop_write:
-		mnt_drop_write(path.mnt);
-		if (err)
-			goto out_mknod_dput;
-		mutex_unlock(&path.dentry->d_inode->i_mutex);
-		dput(path.dentry);
-		path.dentry = dentry;
-
+		err = unix_mknod(sun_path, mode, &path);
+		if (err) {
+			if (err == -EEXIST)
+				err = -EADDRINUSE;
+			unix_release_addr(addr);
+			goto out_up;
+		}
 		addr->hash = UNIX_HASH_SIZE;
-	}
-
-	spin_lock(&unix_table_lock);
-
-	if (!sun_path[0]) {
+		hash = path.dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1);
+		spin_lock(&unix_table_lock);
+		u->path = path;
+		list = &unix_socket_table[hash];
+	} else {
+		spin_lock(&unix_table_lock);
 		err = -EADDRINUSE;
 		if (__unix_find_socket_byname(net, sunaddr, addr_len,
 					      sk->sk_type, hash)) {
···
 		}
 
 		list = &unix_socket_table[addr->hash];
-	} else {
-		list = &unix_socket_table[dentry->d_inode->i_ino & (UNIX_HASH_SIZE-1)];
-		u->path = path;
 	}
 
 	err = 0;
···
 	mutex_unlock(&u->readlock);
 out:
 	return err;
-
-out_mknod_dput:
-	dput(dentry);
-	mutex_unlock(&path.dentry->d_inode->i_mutex);
-	path_put(&path);
-out_mknod_parent:
-	if (err == -EEXIST)
-		err = -EADDRINUSE;
-	unix_release_addr(addr);
-	goto out_up;
 }
 
 static void unix_state_double_lock(struct sock *sk1, struct sock *sk2)

sound/sound_firmware.c

···
 	if (l <= 0 || l > 131072)
 	{
 		printk(KERN_INFO "Invalid firmware '%s'\n", fn);
-		filp_close(filp, current->files);
+		filp_close(filp, NULL);
 		return 0;
 	}
 	dp = vmalloc(l);
 	if (dp == NULL)
 	{
 		printk(KERN_INFO "Out of memory loading '%s'.\n", fn);
-		filp_close(filp, current->files);
+		filp_close(filp, NULL);
 		return 0;
 	}
 	pos = 0;
···
 	{
 		printk(KERN_INFO "Failed to read '%s'.\n", fn);
 		vfree(dp);
-		filp_close(filp, current->files);
+		filp_close(filp, NULL);
 		return 0;
 	}
-	filp_close(filp, current->files);
+	filp_close(filp, NULL);
 	*fp = dp;
 	return (int) l;
 }