[mirror_zfs.git] / module / zfs / zpl_inode.c

/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright (c) 2011, Lawrence Livermore National Security, LLC.
 */


#include <sys/zfs_vfsops.h>
#include <sys/zfs_vnops.h>
#include <sys/zfs_znode.h>
#include <sys/vfs.h>
#include <sys/zpl.h>


static struct dentry *
#ifdef HAVE_LOOKUP_NAMEIDATA
zpl_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
#else
zpl_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags)
#endif
{
	cred_t *cr = CRED();
	struct inode *ip;
	int error;
	fstrans_cookie_t cookie;

	if (dlen(dentry) > ZFS_MAXNAMELEN)
		return (ERR_PTR(-ENAMETOOLONG));

	crhold(cr);
	cookie = spl_fstrans_mark();
	error = -zfs_lookup(dir, dname(dentry), &ip, 0, cr, NULL, NULL);
	spl_fstrans_unmark(cookie);
	ASSERT3S(error, <=, 0);
	crfree(cr);

	spin_lock(&dentry->d_lock);
	dentry->d_time = jiffies;
#ifndef HAVE_S_D_OP
	d_set_d_op(dentry, &zpl_dentry_operations);
#endif /* HAVE_S_D_OP */
	spin_unlock(&dentry->d_lock);

	if (error) {
		if (error == -ENOENT)
			return (d_splice_alias(NULL, dentry));
		else
			return (ERR_PTR(error));
	}

	return (d_splice_alias(ip, dentry));
}

void
zpl_vap_init(vattr_t *vap, struct inode *dir, zpl_umode_t mode, cred_t *cr)
{
	vap->va_mask = ATTR_MODE;
	vap->va_mode = mode;
	vap->va_uid = crgetfsuid(cr);

	if (dir && dir->i_mode & S_ISGID) {
		vap->va_gid = KGID_TO_SGID(dir->i_gid);
		if (S_ISDIR(mode))
			vap->va_mode |= S_ISGID;
	} else {
		vap->va_gid = crgetfsgid(cr);
	}
}

static int
#ifdef HAVE_CREATE_NAMEIDATA
zpl_create(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
    struct nameidata *nd)
#else
zpl_create(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
    bool flag)
#endif
{
	cred_t *cr = CRED();
	struct inode *ip;
	vattr_t *vap;
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
	zpl_vap_init(vap, dir, mode, cr);

	cookie = spl_fstrans_mark();
	error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
	spl_fstrans_unmark(cookie);
	if (error == 0) {
		VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
		VERIFY0(zpl_init_acl(ip, dir));
		d_instantiate(dentry, ip);
	}

	kmem_free(vap, sizeof (vattr_t));
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_mknod(struct inode *dir, struct dentry *dentry, zpl_umode_t mode,
    dev_t rdev)
{
	cred_t *cr = CRED();
	struct inode *ip;
	vattr_t *vap;
	int error;
	fstrans_cookie_t cookie;

	/*
	 * We currently expect Linux to supply rdev=0 for all sockets
	 * and fifos, but we want to know if this behavior ever changes.
	 */
	if (S_ISSOCK(mode) || S_ISFIFO(mode))
		ASSERT(rdev == 0);

	crhold(cr);
	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
	zpl_vap_init(vap, dir, mode, cr);
	vap->va_rdev = rdev;

	cookie = spl_fstrans_mark();
	error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
	spl_fstrans_unmark(cookie);
	if (error == 0) {
		VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
		VERIFY0(zpl_init_acl(ip, dir));
		d_instantiate(dentry, ip);
	}

	kmem_free(vap, sizeof (vattr_t));
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_unlink(struct inode *dir, struct dentry *dentry)
{
	cred_t *cr = CRED();
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	cookie = spl_fstrans_mark();
	error = -zfs_remove(dir, dname(dentry), cr);
	spl_fstrans_unmark(cookie);
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_mkdir(struct inode *dir, struct dentry *dentry, zpl_umode_t mode)
{
	cred_t *cr = CRED();
	vattr_t *vap;
	struct inode *ip;
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
	zpl_vap_init(vap, dir, mode | S_IFDIR, cr);

	cookie = spl_fstrans_mark();
	error = -zfs_mkdir(dir, dname(dentry), vap, &ip, cr, 0, NULL);
	spl_fstrans_unmark(cookie);
	if (error == 0) {
		VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
		VERIFY0(zpl_init_acl(ip, dir));
		d_instantiate(dentry, ip);
	}

	kmem_free(vap, sizeof (vattr_t));
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_rmdir(struct inode * dir, struct dentry *dentry)
{
	cred_t *cr = CRED();
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	cookie = spl_fstrans_mark();
	error = -zfs_rmdir(dir, dname(dentry), NULL, cr, 0);
	spl_fstrans_unmark(cookie);
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
{
	boolean_t issnap = ITOZSB(dentry->d_inode)->z_issnap;
	int error;
	fstrans_cookie_t cookie;

	/*
	 * Ensure MNT_SHRINKABLE is set on snapshots to ensure they are
	 * unmounted automatically with the parent file system.  This
	 * is done on the first getattr because it's not easy to get the
	 * vfsmount structure at mount time.  This call path is explicitly
	 * marked unlikely to avoid any performance impact.  FWIW, ext4
	 * resorts to a similar trick for sysadmin convenience.
	 */
	if (unlikely(issnap && !(mnt->mnt_flags & MNT_SHRINKABLE)))
		mnt->mnt_flags |= MNT_SHRINKABLE;

	cookie = spl_fstrans_mark();
	error = -zfs_getattr_fast(dentry->d_inode, stat);
	spl_fstrans_unmark(cookie);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_setattr(struct dentry *dentry, struct iattr *ia)
{
	struct inode *ip = dentry->d_inode;
	cred_t *cr = CRED();
	vattr_t *vap;
	int error;
	fstrans_cookie_t cookie;

	error = inode_change_ok(ip, ia);
	if (error)
		return (error);

	crhold(cr);
	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
	vap->va_mask = ia->ia_valid & ATTR_IATTR_MASK;
	vap->va_mode = ia->ia_mode;
	vap->va_uid = KUID_TO_SUID(ia->ia_uid);
	vap->va_gid = KGID_TO_SGID(ia->ia_gid);
	vap->va_size = ia->ia_size;
	vap->va_atime = ia->ia_atime;
	vap->va_mtime = ia->ia_mtime;
	vap->va_ctime = ia->ia_ctime;

	cookie = spl_fstrans_mark();
	error = -zfs_setattr(ip, vap, 0, cr);
	spl_fstrans_unmark(cookie);
	if (!error && (ia->ia_valid & ATTR_MODE))
		error = zpl_chmod_acl(ip);

	kmem_free(vap, sizeof (vattr_t));
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_rename(struct inode *sdip, struct dentry *sdentry,
    struct inode *tdip, struct dentry *tdentry)
{
	cred_t *cr = CRED();
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	cookie = spl_fstrans_mark();
	error = -zfs_rename(sdip, dname(sdentry), tdip, dname(tdentry), cr, 0);
	spl_fstrans_unmark(cookie);
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static int
zpl_symlink(struct inode *dir, struct dentry *dentry, const char *name)
{
	cred_t *cr = CRED();
	vattr_t *vap;
	struct inode *ip;
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);
	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
	zpl_vap_init(vap, dir, S_IFLNK | S_IRWXUGO, cr);

	cookie = spl_fstrans_mark();
	error = -zfs_symlink(dir, dname(dentry), vap, (char *)name, &ip, cr, 0);
	spl_fstrans_unmark(cookie);
	if (error == 0) {
		VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
		d_instantiate(dentry, ip);
	}

	kmem_free(vap, sizeof (vattr_t));
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

static void *
zpl_follow_link(struct dentry *dentry, struct nameidata *nd)
{
	cred_t *cr = CRED();
	struct inode *ip = dentry->d_inode;
	struct iovec iov;
	uio_t uio;
	char *link;
	int error;
	fstrans_cookie_t cookie;

	crhold(cr);

	iov.iov_len = MAXPATHLEN;
	iov.iov_base = link = kmem_zalloc(MAXPATHLEN, KM_SLEEP);

	uio.uio_iov = &iov;
	uio.uio_iovcnt = 1;
	uio.uio_resid = (MAXPATHLEN - 1);
	uio.uio_segflg = UIO_SYSSPACE;

	cookie = spl_fstrans_mark();
	error = -zfs_readlink(ip, &uio, cr);
	spl_fstrans_unmark(cookie);
	if (error) {
		kmem_free(link, MAXPATHLEN);
		nd_set_link(nd, ERR_PTR(error));
	} else {
		nd_set_link(nd, link);
	}

	crfree(cr);
	return (NULL);
}

static void
zpl_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr)
{
	const char *link = nd_get_link(nd);

	if (!IS_ERR(link))
		kmem_free(link, MAXPATHLEN);
}

static int
zpl_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
{
	cred_t *cr = CRED();
	struct inode *ip = old_dentry->d_inode;
	int error;
	fstrans_cookie_t cookie;

	if (ip->i_nlink >= ZFS_LINK_MAX)
		return (-EMLINK);

	crhold(cr);
	ip->i_ctime = CURRENT_TIME_SEC;
	igrab(ip); /* Use ihold() if available */

	cookie = spl_fstrans_mark();
	error = -zfs_link(dir, ip, dname(dentry), cr);
	spl_fstrans_unmark(cookie);
	if (error) {
		iput(ip);
		goto out;
	}

	d_instantiate(dentry, ip);
out:
	crfree(cr);
	ASSERT3S(error, <=, 0);

	return (error);
}

#ifdef HAVE_INODE_TRUNCATE_RANGE
static void
zpl_truncate_range(struct inode *ip, loff_t start, loff_t end)
{
	cred_t *cr = CRED();
	flock64_t bf;
	fstrans_cookie_t cookie;

	ASSERT3S(start, <=, end);

	/*
	 * zfs_freesp() will interpret (len == 0) as meaning "truncate until
	 * the end of the file". We don't want that.
	 */
	if (start == end)
		return;

	crhold(cr);

	bf.l_type = F_WRLCK;
	bf.l_whence = 0;
	bf.l_start = start;
	bf.l_len = end - start;
	bf.l_pid = 0;
	cookie = spl_fstrans_mark();
	zfs_space(ip, F_FREESP, &bf, FWRITE, start, cr);
	spl_fstrans_unmark(cookie);

	crfree(cr);
}
#endif /* HAVE_INODE_TRUNCATE_RANGE */

#ifdef HAVE_INODE_FALLOCATE
static long
zpl_fallocate(struct inode *ip, int mode, loff_t offset, loff_t len)
{
	return (zpl_fallocate_common(ip, mode, offset, len));
}
#endif /* HAVE_INODE_FALLOCATE */

static int
#ifdef HAVE_D_REVALIDATE_NAMEIDATA
zpl_revalidate(struct dentry *dentry, struct nameidata *nd)
{
	unsigned int flags = (nd ? nd->flags : 0);
#else
zpl_revalidate(struct dentry *dentry, unsigned int flags)
{
#endif /* HAVE_D_REVALIDATE_NAMEIDATA */
	zfs_sb_t *zsb = dentry->d_sb->s_fs_info;
	int error;

	if (flags & LOOKUP_RCU)
		return (-ECHILD);

	/*
	 * After a rollback negative dentries created before the rollback
	 * time must be invalidated.  Otherwise they can obscure files which
	 * are only present in the rolled back dataset.
	 */
	if (dentry->d_inode == NULL) {
		spin_lock(&dentry->d_lock);
		error = time_before(dentry->d_time, zsb->z_rollback_time);
		spin_unlock(&dentry->d_lock);

		if (error)
			return (0);
	}

	/*
	 * The dentry may reference a stale inode if a mounted file system
	 * was rolled back to a point in time where the object didn't exist.
	 */
	if (dentry->d_inode && ITOZ(dentry->d_inode)->z_is_stale)
		return (0);

	return (1);
}

const struct inode_operations zpl_inode_operations = {
	.create		= zpl_create,
	.link		= zpl_link,
	.unlink		= zpl_unlink,
	.symlink	= zpl_symlink,
	.mkdir		= zpl_mkdir,
	.rmdir		= zpl_rmdir,
	.mknod		= zpl_mknod,
	.rename		= zpl_rename,
	.setattr	= zpl_setattr,
	.getattr	= zpl_getattr,
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.removexattr	= generic_removexattr,
	.listxattr	= zpl_xattr_list,
#ifdef HAVE_INODE_TRUNCATE_RANGE
	.truncate_range = zpl_truncate_range,
#endif /* HAVE_INODE_TRUNCATE_RANGE */
#ifdef HAVE_INODE_FALLOCATE
	.fallocate	= zpl_fallocate,
#endif /* HAVE_INODE_FALLOCATE */
#if defined(CONFIG_FS_POSIX_ACL)
#if defined(HAVE_GET_ACL)
	.get_acl	= zpl_get_acl,
#elif defined(HAVE_CHECK_ACL)
	.check_acl	= zpl_check_acl,
#elif defined(HAVE_PERMISSION)
	.permission	= zpl_permission,
#endif /* HAVE_GET_ACL | HAVE_CHECK_ACL | HAVE_PERMISSION */
#endif /* CONFIG_FS_POSIX_ACL */
};

const struct inode_operations zpl_dir_inode_operations = {
	.create		= zpl_create,
	.lookup		= zpl_lookup,
	.link		= zpl_link,
	.unlink		= zpl_unlink,
	.symlink	= zpl_symlink,
	.mkdir		= zpl_mkdir,
	.rmdir		= zpl_rmdir,
	.mknod		= zpl_mknod,
	.rename		= zpl_rename,
	.setattr	= zpl_setattr,
	.getattr	= zpl_getattr,
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.removexattr	= generic_removexattr,
	.listxattr	= zpl_xattr_list,
#if defined(CONFIG_FS_POSIX_ACL)
#if defined(HAVE_GET_ACL)
	.get_acl	= zpl_get_acl,
#elif defined(HAVE_CHECK_ACL)
	.check_acl	= zpl_check_acl,
#elif defined(HAVE_PERMISSION)
	.permission	= zpl_permission,
#endif /* HAVE_GET_ACL | HAVE_CHECK_ACL | HAVE_PERMISSION */
#endif /* CONFIG_FS_POSIX_ACL */
};

const struct inode_operations zpl_symlink_inode_operations = {
	.readlink	= generic_readlink,
	.follow_link	= zpl_follow_link,
	.put_link	= zpl_put_link,
	.setattr	= zpl_setattr,
	.getattr	= zpl_getattr,
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.removexattr	= generic_removexattr,
	.listxattr	= zpl_xattr_list,
};

const struct inode_operations zpl_special_inode_operations = {
	.setattr	= zpl_setattr,
	.getattr	= zpl_getattr,
	.setxattr	= generic_setxattr,
	.getxattr	= generic_getxattr,
	.removexattr	= generic_removexattr,
	.listxattr	= zpl_xattr_list,
#if defined(CONFIG_FS_POSIX_ACL)
#if defined(HAVE_GET_ACL)
	.get_acl	= zpl_get_acl,
#elif defined(HAVE_CHECK_ACL)
	.check_acl	= zpl_check_acl,
#elif defined(HAVE_PERMISSION)
	.permission	= zpl_permission,
#endif /* HAVE_GET_ACL | HAVE_CHECK_ACL | HAVE_PERMISSION */
#endif /* CONFIG_FS_POSIX_ACL */
};

dentry_operations_t zpl_dentry_operations = {
	.d_revalidate	= zpl_revalidate,
};
Commit	Line	Data
ee154f01 BB	1	/*
	2	* CDDL HEADER START
	3	*
	4	* The contents of this file are subject to the terms of the
	5	* Common Development and Distribution License (the "License").
	6	* You may not use this file except in compliance with the License.
	7	*
	8	* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
	9	* or http://www.opensolaris.org/os/licensing.
	10	* See the License for the specific language governing permissions
	11	* and limitations under the License.
	12	*
	13	* When distributing Covered Code, include this CDDL HEADER in each
	14	* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
	15	* If applicable, add the following below this CDDL HEADER, with the
	16	* fields enclosed by brackets "[]" replaced with your own identifying
	17	* information: Portions Copyright [yyyy] [name of copyright owner]
	18	*
	19	* CDDL HEADER END
	20	*/
	21	/*
	22	* Copyright (c) 2011, Lawrence Livermore National Security, LLC.
	23	*/
	24
	25
	26	#include <sys/zfs_vfsops.h>
	27	#include <sys/zfs_vnops.h>
ebe7e575	28	#include <sys/zfs_znode.h>
ee154f01 BB	29	#include <sys/vfs.h>
	30	#include <sys/zpl.h>
	31
	32
	33	static struct dentry *
8f195a90	34	#ifdef HAVE_LOOKUP_NAMEIDATA
ee154f01	35	zpl_lookup(struct inode dir, struct dentry dentry, struct nameidata *nd)
8f195a90 YS	36	#else
	37	zpl_lookup(struct inode dir, struct dentry dentry, unsigned int flags)
	38	#endif
ee154f01	39	{
81e97e21	40	cred_t *cr = CRED();
ee154f01	41	struct inode *ip;
ee154f01	42	int error;
40d06e3c	43	fstrans_cookie_t cookie;
ee154f01	44
9878a89d	45	if (dlen(dentry) > ZFS_MAXNAMELEN)
d1d7e268	46	return (ERR_PTR(-ENAMETOOLONG));
9878a89d	47
81e97e21	48	crhold(cr);
40d06e3c	49	cookie = spl_fstrans_mark();
ee154f01	50	error = -zfs_lookup(dir, dname(dentry), &ip, 0, cr, NULL, NULL);
40d06e3c	51	spl_fstrans_unmark(cookie);
ee154f01	52	ASSERT3S(error, <=, 0);
81e97e21	53	crfree(cr);
ee154f01	54
7b3e34ba BB	55	spin_lock(&dentry->d_lock);
7b3e34ba BB	56	dentry->d_time = jiffies;
ee930353 BB	57	#ifndef HAVE_S_D_OP
	58	d_set_d_op(dentry, &zpl_dentry_operations);
	59	#endif /* HAVE_S_D_OP */
7b3e34ba BB	60	spin_unlock(&dentry->d_lock);
7b3e34ba BB	61
ee154f01 BB	62	if (error) {
ee154f01 BB	63	if (error == -ENOENT)
d1d7e268	64	return (d_splice_alias(NULL, dentry));
ee154f01	65	else
d1d7e268	66	return (ERR_PTR(error));
ee154f01 BB	67	}
ee154f01 BB	68
d1d7e268	69	return (d_splice_alias(ip, dentry));
ee154f01 BB	70	}
ee154f01 BB	71
ebe7e575	72	void
7b3e34ba	73	zpl_vap_init(vattr_t vap, struct inode dir, zpl_umode_t mode, cred_t *cr)
9fd91dae BB	74	{
	75	vap->va_mask = ATTR_MODE;
	76	vap->va_mode = mode;
9fd91dae BB	77	vap->va_uid = crgetfsuid(cr);
	78
	79	if (dir && dir->i_mode & S_ISGID) {
570d6edf	80	vap->va_gid = KGID_TO_SGID(dir->i_gid);
9fd91dae BB	81	if (S_ISDIR(mode))
	82	vap->va_mode \|= S_ISGID;
	83	} else {
	84	vap->va_gid = crgetfsgid(cr);
	85	}
	86	}
	87
ee154f01	88	static int
558ef6d0	89	#ifdef HAVE_CREATE_NAMEIDATA
b39d3b9f	90	zpl_create(struct inode dir, struct dentry dentry, zpl_umode_t mode,
ee154f01	91	struct nameidata *nd)
558ef6d0 YS	92	#else
	93	zpl_create(struct inode dir, struct dentry dentry, zpl_umode_t mode,
	94	bool flag)
	95	#endif
ee154f01	96	{
81e97e21	97	cred_t *cr = CRED();
ee154f01 BB	98	struct inode *ip;
	99	vattr_t *vap;
	100	int error;
40d06e3c	101	fstrans_cookie_t cookie;
ee154f01	102
81e97e21	103	crhold(cr);
d1d7e268	104	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
7b3e34ba	105	zpl_vap_init(vap, dir, mode, cr);
ee154f01	106
40d06e3c	107	cookie = spl_fstrans_mark();
e89260a1	108	error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
40d06e3c	109	spl_fstrans_unmark(cookie);
e89260a1	110	if (error == 0) {
023699cd MM	111	VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
023699cd MM	112	VERIFY0(zpl_init_acl(ip, dir));
7b3e34ba	113	d_instantiate(dentry, ip);
e89260a1 BB	114	}
e89260a1 BB	115
d1d7e268	116	kmem_free(vap, sizeof (vattr_t));
81e97e21	117	crfree(cr);
ee154f01 BB	118	ASSERT3S(error, <=, 0);
	119
	120	return (error);
	121	}
	122
	123	static int
b39d3b9f BB	124	zpl_mknod(struct inode dir, struct dentry dentry, zpl_umode_t mode,
b39d3b9f BB	125	dev_t rdev)
ee154f01	126	{
81e97e21	127	cred_t *cr = CRED();
ee154f01 BB	128	struct inode *ip;
	129	vattr_t *vap;
	130	int error;
40d06e3c	131	fstrans_cookie_t cookie;
ee154f01	132
aa6d8c10 NB	133	/*
	134	* We currently expect Linux to supply rdev=0 for all sockets
	135	* and fifos, but we want to know if this behavior ever changes.
	136	*/
	137	if (S_ISSOCK(mode) \|\| S_ISFIFO(mode))
	138	ASSERT(rdev == 0);
	139
81e97e21	140	crhold(cr);
d1d7e268	141	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
7b3e34ba	142	zpl_vap_init(vap, dir, mode, cr);
ee154f01	143	vap->va_rdev = rdev;
ee154f01	144
40d06e3c	145	cookie = spl_fstrans_mark();
7b3e34ba	146	error = -zfs_create(dir, dname(dentry), vap, 0, mode, &ip, cr, 0, NULL);
40d06e3c	147	spl_fstrans_unmark(cookie);
023699cd	148	if (error == 0) {
227bc969	149	VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
023699cd	150	VERIFY0(zpl_init_acl(ip, dir));
7b3e34ba	151	d_instantiate(dentry, ip);
023699cd	152	}
7b3e34ba	153
d1d7e268	154	kmem_free(vap, sizeof (vattr_t));
81e97e21	155	crfree(cr);
ee154f01 BB	156	ASSERT3S(error, <=, 0);
ee154f01 BB	157
34d5a5fd	158	return (error);
ee154f01 BB	159	}
	160
	161	static int
	162	zpl_unlink(struct inode dir, struct dentry dentry)
	163	{
81e97e21	164	cred_t *cr = CRED();
ee154f01	165	int error;
40d06e3c	166	fstrans_cookie_t cookie;
ee154f01	167
81e97e21	168	crhold(cr);
40d06e3c	169	cookie = spl_fstrans_mark();
ee154f01	170	error = -zfs_remove(dir, dname(dentry), cr);
40d06e3c	171	spl_fstrans_unmark(cookie);
81e97e21	172	crfree(cr);
ee154f01 BB	173	ASSERT3S(error, <=, 0);
	174
	175	return (error);
	176	}
	177
	178	static int
b39d3b9f	179	zpl_mkdir(struct inode dir, struct dentry dentry, zpl_umode_t mode)
ee154f01	180	{
81e97e21	181	cred_t *cr = CRED();
ee154f01 BB	182	vattr_t *vap;
	183	struct inode *ip;
	184	int error;
40d06e3c	185	fstrans_cookie_t cookie;
ee154f01	186
81e97e21	187	crhold(cr);
d1d7e268	188	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
7b3e34ba	189	zpl_vap_init(vap, dir, mode \| S_IFDIR, cr);
ee154f01	190
40d06e3c	191	cookie = spl_fstrans_mark();
ee154f01	192	error = -zfs_mkdir(dir, dname(dentry), vap, &ip, cr, 0, NULL);
40d06e3c	193	spl_fstrans_unmark(cookie);
023699cd	194	if (error == 0) {
227bc969	195	VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
023699cd	196	VERIFY0(zpl_init_acl(ip, dir));
7b3e34ba	197	d_instantiate(dentry, ip);
023699cd	198	}
7b3e34ba	199
d1d7e268	200	kmem_free(vap, sizeof (vattr_t));
81e97e21	201	crfree(cr);
ee154f01 BB	202	ASSERT3S(error, <=, 0);
	203
	204	return (error);
	205	}
	206
	207	static int
	208	zpl_rmdir(struct inode * dir, struct dentry *dentry)
	209	{
81e97e21	210	cred_t *cr = CRED();
ee154f01	211	int error;
40d06e3c	212	fstrans_cookie_t cookie;
ee154f01	213
81e97e21	214	crhold(cr);
40d06e3c	215	cookie = spl_fstrans_mark();
ee154f01	216	error = -zfs_rmdir(dir, dname(dentry), NULL, cr, 0);
40d06e3c	217	spl_fstrans_unmark(cookie);
81e97e21	218	crfree(cr);
ee154f01 BB	219	ASSERT3S(error, <=, 0);
	220
	221	return (error);
	222	}
	223
	224	static int
	225	zpl_getattr(struct vfsmount mnt, struct dentry dentry, struct kstat *stat)
	226	{
ebe7e575	227	boolean_t issnap = ITOZSB(dentry->d_inode)->z_issnap;
ee154f01	228	int error;
40d06e3c	229	fstrans_cookie_t cookie;
ee154f01	230
ebe7e575 BB	231	/*
	232	* Ensure MNT_SHRINKABLE is set on snapshots to ensure they are
	233	* unmounted automatically with the parent file system. This
	234	* is done on the first getattr because it's not easy to get the
	235	* vfsmount structure at mount time. This call path is explicitly
	236	* marked unlikely to avoid any performance impact. FWIW, ext4
	237	* resorts to a similar trick for sysadmin convenience.
	238	*/
	239	if (unlikely(issnap && !(mnt->mnt_flags & MNT_SHRINKABLE)))
	240	mnt->mnt_flags \|= MNT_SHRINKABLE;
	241
40d06e3c	242	cookie = spl_fstrans_mark();
057e8eee	243	error = -zfs_getattr_fast(dentry->d_inode, stat);
40d06e3c	244	spl_fstrans_unmark(cookie);
ee154f01 BB	245	ASSERT3S(error, <=, 0);
	246
	247	return (error);
	248	}
	249
	250	static int
5484965a	251	zpl_setattr(struct dentry dentry, struct iattr ia)
ee154f01	252	{
023699cd	253	struct inode *ip = dentry->d_inode;
81e97e21	254	cred_t *cr = CRED();
5484965a	255	vattr_t *vap;
ee154f01	256	int error;
40d06e3c	257	fstrans_cookie_t cookie;
ee154f01	258
023699cd	259	error = inode_change_ok(ip, ia);
ee154f01 BB	260	if (error)
	261	return (error);
	262
81e97e21	263	crhold(cr);
d1d7e268	264	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
5484965a BB	265	vap->va_mask = ia->ia_valid & ATTR_IATTR_MASK;
5484965a BB	266	vap->va_mode = ia->ia_mode;
570d6edf RY	267	vap->va_uid = KUID_TO_SUID(ia->ia_uid);
570d6edf RY	268	vap->va_gid = KGID_TO_SGID(ia->ia_gid);
5484965a BB	269	vap->va_size = ia->ia_size;
	270	vap->va_atime = ia->ia_atime;
	271	vap->va_mtime = ia->ia_mtime;
	272	vap->va_ctime = ia->ia_ctime;
	273
40d06e3c	274	cookie = spl_fstrans_mark();
023699cd	275	error = -zfs_setattr(ip, vap, 0, cr);
40d06e3c	276	spl_fstrans_unmark(cookie);
023699cd MM	277	if (!error && (ia->ia_valid & ATTR_MODE))
023699cd MM	278	error = zpl_chmod_acl(ip);
5484965a	279
d1d7e268	280	kmem_free(vap, sizeof (vattr_t));
81e97e21	281	crfree(cr);
ee154f01 BB	282	ASSERT3S(error, <=, 0);
ee154f01 BB	283
5484965a	284	return (error);
ee154f01 BB	285	}
	286
	287	static int
	288	zpl_rename(struct inode sdip, struct dentry sdentry,
	289	struct inode tdip, struct dentry tdentry)
	290	{
81e97e21	291	cred_t *cr = CRED();
ee154f01	292	int error;
40d06e3c	293	fstrans_cookie_t cookie;
ee154f01	294
81e97e21	295	crhold(cr);
40d06e3c	296	cookie = spl_fstrans_mark();
ee154f01	297	error = -zfs_rename(sdip, dname(sdentry), tdip, dname(tdentry), cr, 0);
40d06e3c	298	spl_fstrans_unmark(cookie);
81e97e21	299	crfree(cr);
ee154f01 BB	300	ASSERT3S(error, <=, 0);
	301
	302	return (error);
	303	}
	304
	305	static int
	306	zpl_symlink(struct inode dir, struct dentry dentry, const char *name)
	307	{
81e97e21	308	cred_t *cr = CRED();
ee154f01 BB	309	vattr_t *vap;
	310	struct inode *ip;
	311	int error;
40d06e3c	312	fstrans_cookie_t cookie;
ee154f01	313
81e97e21	314	crhold(cr);
d1d7e268	315	vap = kmem_zalloc(sizeof (vattr_t), KM_SLEEP);
7b3e34ba	316	zpl_vap_init(vap, dir, S_IFLNK \| S_IRWXUGO, cr);
ee154f01	317
40d06e3c	318	cookie = spl_fstrans_mark();
ee154f01	319	error = -zfs_symlink(dir, dname(dentry), vap, (char *)name, &ip, cr, 0);
40d06e3c	320	spl_fstrans_unmark(cookie);
227bc969 MT	321	if (error == 0) {
227bc969 MT	322	VERIFY0(zpl_xattr_security_init(ip, dir, &dentry->d_name));
7b3e34ba	323	d_instantiate(dentry, ip);
227bc969	324	}
7b3e34ba	325
d1d7e268	326	kmem_free(vap, sizeof (vattr_t));
81e97e21	327	crfree(cr);
ee154f01 BB	328	ASSERT3S(error, <=, 0);
	329
	330	return (error);
	331	}
	332
	333	static void *
	334	zpl_follow_link(struct dentry dentry, struct nameidata nd)
	335	{
81e97e21	336	cred_t *cr = CRED();
8b4f9a2d BB	337	struct inode *ip = dentry->d_inode;
	338	struct iovec iov;
	339	uio_t uio;
	340	char *link;
8b4f9a2d	341	int error;
40d06e3c	342	fstrans_cookie_t cookie;
8b4f9a2d	343
81e97e21	344	crhold(cr);
8b4f9a2d BB	345
	346	iov.iov_len = MAXPATHLEN;
	347	iov.iov_base = link = kmem_zalloc(MAXPATHLEN, KM_SLEEP);
	348
	349	uio.uio_iov = &iov;
	350	uio.uio_iovcnt = 1;
	351	uio.uio_resid = (MAXPATHLEN - 1);
	352	uio.uio_segflg = UIO_SYSSPACE;
	353
40d06e3c	354	cookie = spl_fstrans_mark();
50950001	355	error = -zfs_readlink(ip, &uio, cr);
40d06e3c	356	spl_fstrans_unmark(cookie);
8b4f9a2d BB	357	if (error) {
	358	kmem_free(link, MAXPATHLEN);
	359	nd_set_link(nd, ERR_PTR(error));
	360	} else {
	361	nd_set_link(nd, link);
	362	}
	363
81e97e21	364	crfree(cr);
8b4f9a2d	365	return (NULL);
ee154f01 BB	366	}
	367
	368	static void
	369	zpl_put_link(struct dentry dentry, struct nameidata nd, void *ptr)
	370	{
0a6b03d3	371	const char *link = nd_get_link(nd);
ee154f01	372
ee154f01	373	if (!IS_ERR(link))
8b4f9a2d	374	kmem_free(link, MAXPATHLEN);
ee154f01 BB	375	}
	376
	377	static int
	378	zpl_link(struct dentry old_dentry, struct inode dir, struct dentry *dentry)
	379	{
81e97e21	380	cred_t *cr = CRED();
ee154f01	381	struct inode *ip = old_dentry->d_inode;
ee154f01	382	int error;
40d06e3c	383	fstrans_cookie_t cookie;
ee154f01 BB	384
ee154f01 BB	385	if (ip->i_nlink >= ZFS_LINK_MAX)
d1d7e268	386	return (-EMLINK);
ee154f01	387
81e97e21	388	crhold(cr);
ee154f01 BB	389	ip->i_ctime = CURRENT_TIME_SEC;
	390	igrab(ip); /* Use ihold() if available */
	391
40d06e3c	392	cookie = spl_fstrans_mark();
ee154f01	393	error = -zfs_link(dir, ip, dname(dentry), cr);
40d06e3c	394	spl_fstrans_unmark(cookie);
ee154f01 BB	395	if (error) {
	396	iput(ip);
	397	goto out;
	398	}
	399
	400	d_instantiate(dentry, ip);
	401	out:
81e97e21	402	crfree(cr);
ee154f01 BB	403	ASSERT3S(error, <=, 0);
	404
	405	return (error);
	406	}
	407
ea1fdf46	408	#ifdef HAVE_INODE_TRUNCATE_RANGE
5cb63a57	409	static void
d1d7e268	410	zpl_truncate_range(struct inode *ip, loff_t start, loff_t end)
5cb63a57 ED	411	{
	412	cred_t *cr = CRED();
	413	flock64_t bf;
40d06e3c	414	fstrans_cookie_t cookie;
5cb63a57 ED	415
	416	ASSERT3S(start, <=, end);
	417
	418	/*
	419	* zfs_freesp() will interpret (len == 0) as meaning "truncate until
	420	* the end of the file". We don't want that.
	421	*/
	422	if (start == end)
	423	return;
	424
	425	crhold(cr);
	426
	427	bf.l_type = F_WRLCK;
	428	bf.l_whence = 0;
	429	bf.l_start = start;
	430	bf.l_len = end - start;
	431	bf.l_pid = 0;
40d06e3c	432	cookie = spl_fstrans_mark();
5cb63a57	433	zfs_space(ip, F_FREESP, &bf, FWRITE, start, cr);
40d06e3c	434	spl_fstrans_unmark(cookie);
5cb63a57 ED	435
	436	crfree(cr);
	437	}
ea1fdf46	438	#endif /* HAVE_INODE_TRUNCATE_RANGE */
5cb63a57	439
cb2d1901 ED	440	#ifdef HAVE_INODE_FALLOCATE
	441	static long
	442	zpl_fallocate(struct inode *ip, int mode, loff_t offset, loff_t len)
	443	{
d1d7e268	444	return (zpl_fallocate_common(ip, mode, offset, len));
cb2d1901 ED	445	}
	446	#endif /* HAVE_INODE_FALLOCATE */
	447
7b3e34ba BB	448	static int
	449	#ifdef HAVE_D_REVALIDATE_NAMEIDATA
	450	zpl_revalidate(struct dentry dentry, struct nameidata nd)
	451	{
09a661e9	452	unsigned int flags = (nd ? nd->flags : 0);
7b3e34ba BB	453	#else
	454	zpl_revalidate(struct dentry *dentry, unsigned int flags)
	455	{
	456	#endif /* HAVE_D_REVALIDATE_NAMEIDATA */
	457	zfs_sb_t *zsb = dentry->d_sb->s_fs_info;
	458	int error;
	459
	460	if (flags & LOOKUP_RCU)
	461	return (-ECHILD);
	462
	463	/*
	464	* After a rollback negative dentries created before the rollback
	465	* time must be invalidated. Otherwise they can obscure files which
	466	* are only present in the rolled back dataset.
	467	*/
	468	if (dentry->d_inode == NULL) {
	469	spin_lock(&dentry->d_lock);
	470	error = time_before(dentry->d_time, zsb->z_rollback_time);
	471	spin_unlock(&dentry->d_lock);
	472
	473	if (error)
	474	return (0);
	475	}
	476
	477	/*
	478	* The dentry may reference a stale inode if a mounted file system
	479	* was rolled back to a point in time where the object didn't exist.
	480	*/
	481	if (dentry->d_inode && ITOZ(dentry->d_inode)->z_is_stale)
	482	return (0);
	483
	484	return (1);
	485	}
cb2d1901	486
ee154f01	487	const struct inode_operations zpl_inode_operations = {
ee154f01 BB	488	.create = zpl_create,
	489	.link = zpl_link,
	490	.unlink = zpl_unlink,
	491	.symlink = zpl_symlink,
	492	.mkdir = zpl_mkdir,
	493	.rmdir = zpl_rmdir,
	494	.mknod = zpl_mknod,
	495	.rename = zpl_rename,
	496	.setattr = zpl_setattr,
	497	.getattr = zpl_getattr,
	498	.setxattr = generic_setxattr,
	499	.getxattr = generic_getxattr,
	500	.removexattr = generic_removexattr,
	501	.listxattr = zpl_xattr_list,
ea1fdf46	502	#ifdef HAVE_INODE_TRUNCATE_RANGE
5cb63a57	503	.truncate_range = zpl_truncate_range,
ea1fdf46	504	#endif /* HAVE_INODE_TRUNCATE_RANGE */
cb2d1901 ED	505	#ifdef HAVE_INODE_FALLOCATE
	506	.fallocate = zpl_fallocate,
	507	#endif /* HAVE_INODE_FALLOCATE */
b695c34e	508	#if defined(CONFIG_FS_POSIX_ACL)
023699cd MM	509	#if defined(HAVE_GET_ACL)
	510	.get_acl = zpl_get_acl,
	511	#elif defined(HAVE_CHECK_ACL)
	512	.check_acl = zpl_check_acl,
	513	#elif defined(HAVE_PERMISSION)
	514	.permission = zpl_permission,
	515	#endif /* HAVE_GET_ACL \| HAVE_CHECK_ACL \| HAVE_PERMISSION */
b695c34e	516	#endif /* CONFIG_FS_POSIX_ACL */
ee154f01 BB	517	};
	518
	519	const struct inode_operations zpl_dir_inode_operations = {
ee154f01 BB	520	.create = zpl_create,
	521	.lookup = zpl_lookup,
	522	.link = zpl_link,
	523	.unlink = zpl_unlink,
	524	.symlink = zpl_symlink,
	525	.mkdir = zpl_mkdir,
	526	.rmdir = zpl_rmdir,
	527	.mknod = zpl_mknod,
	528	.rename = zpl_rename,
	529	.setattr = zpl_setattr,
a6695d83 BB	530	.getattr = zpl_getattr,
	531	.setxattr = generic_setxattr,
	532	.getxattr = generic_getxattr,
	533	.removexattr = generic_removexattr,
	534	.listxattr = zpl_xattr_list,
b695c34e	535	#if defined(CONFIG_FS_POSIX_ACL)
023699cd MM	536	#if defined(HAVE_GET_ACL)
	537	.get_acl = zpl_get_acl,
	538	#elif defined(HAVE_CHECK_ACL)
	539	.check_acl = zpl_check_acl,
	540	#elif defined(HAVE_PERMISSION)
	541	.permission = zpl_permission,
	542	#endif /* HAVE_GET_ACL \| HAVE_CHECK_ACL \| HAVE_PERMISSION */
b695c34e	543	#endif /* CONFIG_FS_POSIX_ACL */
ee154f01 BB	544	};
	545
	546	const struct inode_operations zpl_symlink_inode_operations = {
ee154f01 BB	547	.readlink = generic_readlink,
	548	.follow_link = zpl_follow_link,
	549	.put_link = zpl_put_link,
6f2255ba BB	550	.setattr = zpl_setattr,
6f2255ba BB	551	.getattr = zpl_getattr,
f31b3ebe BB	552	.setxattr = generic_setxattr,
	553	.getxattr = generic_getxattr,
	554	.removexattr = generic_removexattr,
	555	.listxattr = zpl_xattr_list,
ee154f01 BB	556	};
	557
	558	const struct inode_operations zpl_special_inode_operations = {
a6695d83 BB	559	.setattr = zpl_setattr,
	560	.getattr = zpl_getattr,
	561	.setxattr = generic_setxattr,
	562	.getxattr = generic_getxattr,
	563	.removexattr = generic_removexattr,
	564	.listxattr = zpl_xattr_list,
b695c34e	565	#if defined(CONFIG_FS_POSIX_ACL)
023699cd MM	566	#if defined(HAVE_GET_ACL)
	567	.get_acl = zpl_get_acl,
	568	#elif defined(HAVE_CHECK_ACL)
	569	.check_acl = zpl_check_acl,
	570	#elif defined(HAVE_PERMISSION)
	571	.permission = zpl_permission,
	572	#endif /* HAVE_GET_ACL \| HAVE_CHECK_ACL \| HAVE_PERMISSION */
b695c34e	573	#endif /* CONFIG_FS_POSIX_ACL */
ee154f01	574	};
7b3e34ba BB	575
	576	dentry_operations_t zpl_dentry_operations = {
	577	.d_revalidate = zpl_revalidate,
	578	};