void
zfs_sa_upgrade_txholds(dmu_tx_t *tx, znode_t *zp)
{
if (!ZTOZSB(zp)->z_use_sa || zp->z_is_sa)
return;
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_TRUE);
if (zfs_external_acl(zp)) {
dmu_tx_hold_free(tx, zfs_external_acl(zp), 0,
DMU_OBJECT_END);
}
}
/*
* Create a new DMU object to hold a zfs znode.
*
* IN: dzp - parent directory for new znode
* vap - file attributes for new znode
* tx - dmu transaction id for zap operations
* cr - credentials of caller
* flag - flags:
* IS_ROOT_NODE - new object will be root
* IS_XATTR - new object is an attribute
* bonuslen - length of bonus buffer
* setaclp - File/Dir initial ACL
* fuidp - Tracks fuid allocation.
*
* OUT: zpp - allocated znode
*
*/
void
zfs_mknode(znode_t *dzp, vattr_t *vap, dmu_tx_t *tx, cred_t *cr,
uint_t flag, znode_t **zpp, zfs_acl_ids_t *acl_ids)
{
uint64_t crtime[2], atime[2], mtime[2], ctime[2];
uint64_t mode, size, links, parent, pflags;
uint64_t dzp_pflags = 0;
uint64_t rdev = 0;
zfs_sb_t *zsb = ZTOZSB(dzp);
dmu_buf_t *db;
timestruc_t now;
uint64_t gen, obj;
int err;
int bonuslen;
sa_handle_t *sa_hdl;
dmu_object_type_t obj_type;
sa_bulk_attr_t *sa_attrs;
int cnt = 0;
zfs_acl_locator_cb_t locate = { 0 };
if (zsb->z_replay) {
obj = vap->va_nodeid;
now = vap->va_ctime; /* see zfs_replay_create() */
gen = vap->va_nblocks; /* ditto */
} else {
obj = 0;
gethrestime(&now);
gen = dmu_tx_get_txg(tx);
}
obj_type = zsb->z_use_sa ? DMU_OT_SA : DMU_OT_ZNODE;
bonuslen = (obj_type == DMU_OT_SA) ?
DN_MAX_BONUSLEN : ZFS_OLD_ZNODE_PHYS_SIZE;
/*
* Create a new DMU object.
*/
/*
* There's currently no mechanism for pre-reading the blocks that will
* be needed to allocate a new object, so we accept the small chance
* that there will be an i/o error and we will fail one of the
* assertions below.
*/
if (S_ISDIR(vap->va_mode)) {
if (zsb->z_replay) {
err = zap_create_claim_norm(zsb->z_os, obj,
zsb->z_norm, DMU_OT_DIRECTORY_CONTENTS,
obj_type, bonuslen, tx);
ASSERT3U(err, ==, 0);
} else {
static int
zpl_xattr_set(struct inode *ip, const char *name, const void *value,
size_t size, int flags)
{
znode_t *zp = ITOZ(ip);
zfs_sb_t *zsb = ZTOZSB(zp);
cred_t *cr = CRED();
int error;
crhold(cr);
rw_enter(&ITOZ(ip)->z_xattr_lock, RW_WRITER);
/*
* Before setting the xattr check to see if it already exists.
* This is done to ensure the following optional flags are honored.
*
* XATTR_CREATE: fail if xattr already exists
* XATTR_REPLACE: fail if xattr does not exist
*/
error = __zpl_xattr_get(ip, name, NULL, 0, cr);
if (error < 0) {
if (error != -ENODATA)
goto out;
if ((error == -ENODATA) && (flags & XATTR_REPLACE))
goto out;
} else {
error = -EEXIST;
if (flags & XATTR_CREATE)
goto out;
}
/* Preferentially store the xattr as a SA for better performance */
if (zsb->z_use_sa && zsb->z_xattr_sa && zp->z_is_sa) {
error = zpl_xattr_set_sa(ip, name, value, size, flags, cr);
if (error == 0)
goto out;
}
error = zpl_xattr_set_dir(ip, name, value, size, flags, cr);
out:
rw_exit(&ITOZ(ip)->z_xattr_lock);
crfree(cr);
ASSERT3S(error, <=, 0);
return (error);
}
void
zfs_znode_delete(znode_t *zp, dmu_tx_t *tx)
{
zfs_sb_t *zsb = ZTOZSB(zp);
objset_t *os = zsb->z_os;
uint64_t obj = zp->z_id;
uint64_t acl_obj = zfs_external_acl(zp);
ZFS_OBJ_HOLD_ENTER(zsb, obj);
if (acl_obj) {
VERIFY(!zp->z_is_sa);
VERIFY(0 == dmu_object_free(os, acl_obj, tx));
}
VERIFY(0 == dmu_object_free(os, obj, tx));
zfs_znode_dmu_fini(zp);
ZFS_OBJ_HOLD_EXIT(zsb, obj);
}
void
zfs_zinactive(znode_t *zp)
{
zfs_sb_t *zsb = ZTOZSB(zp);
uint64_t z_id = zp->z_id;
boolean_t drop_mutex = 0;
ASSERT(zp->z_sa_hdl);
/*
* Don't allow a zfs_zget() while were trying to release this znode.
*
* Linux allows direct memory reclaim which means that any KM_SLEEP
* allocation may trigger inode eviction. This can lead to a deadlock
* through the ->shrink_icache_memory()->evict()->zfs_inactive()->
* zfs_zinactive() call path. To avoid this deadlock the process
* must not reacquire the mutex when it is already holding it.
*/
if (!ZFS_OBJ_HOLD_OWNED(zsb, z_id)) {
ZFS_OBJ_HOLD_ENTER(zsb, z_id);
drop_mutex = 1;
}
mutex_enter(&zp->z_lock);
/*
* If this was the last reference to a file with no links,
* remove the file from the file system.
*/
if (zp->z_unlinked) {
mutex_exit(&zp->z_lock);
if (drop_mutex)
ZFS_OBJ_HOLD_EXIT(zsb, z_id);
zfs_rmnode(zp);
return;
}
mutex_exit(&zp->z_lock);
zfs_znode_dmu_fini(zp);
if (drop_mutex)
ZFS_OBJ_HOLD_EXIT(zsb, z_id);
}
int
zfs_sa_set_xattr(znode_t *zp)
{
zfs_sb_t *zsb = ZTOZSB(zp);
dmu_tx_t *tx;
char *obj;
size_t size;
int error;
ASSERT(RW_WRITE_HELD(&zp->z_xattr_lock));
ASSERT(zp->z_xattr_cached);
ASSERT(zp->z_is_sa);
error = nvlist_size(zp->z_xattr_cached, &size, NV_ENCODE_XDR);
if (error)
goto out;
obj = zio_buf_alloc(size);
error = nvlist_pack(zp->z_xattr_cached, &obj, &size,
NV_ENCODE_XDR, KM_SLEEP);
if (error)
goto out_free;
tx = dmu_tx_create(zsb->z_os);
dmu_tx_hold_sa_create(tx, size);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_TRUE);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
dmu_tx_abort(tx);
} else {
error = sa_update(zp->z_sa_hdl, SA_ZPL_DXATTR(zsb),
obj, size, tx);
if (error)
dmu_tx_abort(tx);
else
dmu_tx_commit(tx);
}
out_free:
zio_buf_free(obj, size);
out:
return (error);
}
int
zfs_sa_set_xattr(znode_t *zp)
{
zfsvfs_t *zfsvfs = ZTOZSB(zp);
dmu_tx_t *tx;
char *obj;
size_t size;
int error;
ASSERT(RW_WRITE_HELD(&zp->z_xattr_lock));
ASSERT(zp->z_xattr_cached);
ASSERT(zp->z_is_sa);
error = nvlist_size(zp->z_xattr_cached, &size, NV_ENCODE_XDR);
if ((error == 0) && (size > SA_ATTR_MAX_LEN))
error = EFBIG;
if (error)
goto out;
obj = vmem_alloc(size, KM_SLEEP);
error = nvlist_pack(zp->z_xattr_cached, &obj, &size,
NV_ENCODE_XDR, KM_SLEEP);
if (error)
goto out_free;
tx = dmu_tx_create(zfsvfs->z_os);
dmu_tx_hold_sa_create(tx, size);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_TRUE);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
dmu_tx_abort(tx);
} else {
VERIFY0(sa_update(zp->z_sa_hdl, SA_ZPL_DXATTR(zfsvfs),
obj, size, tx));
dmu_tx_commit(tx);
}
out_free:
vmem_free(obj, size);
out:
return (error);
}
static int
zpl_xattr_get(struct inode *ip, const char *name, void *value, size_t size)
{
znode_t *zp = ITOZ(ip);
zfs_sb_t *zsb = ZTOZSB(zp);
cred_t *cr = CRED();
fstrans_cookie_t cookie;
int error;
crhold(cr);
cookie = spl_fstrans_mark();
rrm_enter_read(&(zsb)->z_teardown_lock, FTAG);
rw_enter(&zp->z_xattr_lock, RW_READER);
error = __zpl_xattr_get(ip, name, value, size, cr);
rw_exit(&zp->z_xattr_lock);
rrm_exit(&(zsb)->z_teardown_lock, FTAG);
spl_fstrans_unmark(cookie);
crfree(cr);
return (error);
}
/*
* Update the embedded inode given the znode. We should work toward
* eliminating this function as soon as possible by removing values
* which are duplicated between the znode and inode. If the generic
* inode has the correct field it should be used, and the ZFS code
* updated to access the inode. This can be done incrementally.
*/
void
zfs_inode_update(znode_t *zp)
{
zfs_sb_t *zsb;
struct inode *ip;
uint32_t blksize;
uint64_t atime[2], mtime[2], ctime[2];
ASSERT(zp != NULL);
zsb = ZTOZSB(zp);
ip = ZTOI(zp);
/* Skip .zfs control nodes which do not exist on disk. */
if (zfsctl_is_node(ip))
return;
sa_lookup(zp->z_sa_hdl, SA_ZPL_ATIME(zsb), &atime, 16);
sa_lookup(zp->z_sa_hdl, SA_ZPL_MTIME(zsb), &mtime, 16);
sa_lookup(zp->z_sa_hdl, SA_ZPL_CTIME(zsb), &ctime, 16);
spin_lock(&ip->i_lock);
ip->i_generation = zp->z_gen;
ip->i_uid = SUID_TO_KUID(zp->z_uid);
ip->i_gid = SGID_TO_KGID(zp->z_gid);
set_nlink(ip, zp->z_links);
ip->i_mode = zp->z_mode;
zfs_set_inode_flags(zp, ip);
ip->i_blkbits = SPA_MINBLOCKSHIFT;
dmu_object_size_from_db(sa_get_db(zp->z_sa_hdl), &blksize,
(u_longlong_t *)&ip->i_blocks);
ZFS_TIME_DECODE(&ip->i_atime, atime);
ZFS_TIME_DECODE(&ip->i_mtime, mtime);
ZFS_TIME_DECODE(&ip->i_ctime, ctime);
i_size_write(ip, zp->z_size);
spin_unlock(&ip->i_lock);
}
int
zfs_sa_readlink(znode_t *zp, uio_t *uio)
{
dmu_buf_t *db = sa_get_db(zp->z_sa_hdl);
size_t bufsz;
int error;
bufsz = zp->z_size;
if (bufsz + ZFS_OLD_ZNODE_PHYS_SIZE <= db->db_size) {
error = uiomove((caddr_t)db->db_data +
ZFS_OLD_ZNODE_PHYS_SIZE,
MIN((size_t)bufsz, uio->uio_resid), UIO_READ, uio);
} else {
dmu_buf_t *dbp;
if ((error = dmu_buf_hold(ZTOZSB(zp)->z_os, zp->z_id,
0, FTAG, &dbp, DMU_READ_NO_PREFETCH)) == 0) {
error = uiomove(dbp->db_data,
MIN((size_t)bufsz, uio->uio_resid), UIO_READ, uio);
dmu_buf_rele(dbp, FTAG);
}
}
return (error);
}
/*
* If zp is an xattr node, check whether the xattr owner is unlinked.
* We don't want to log anything if the owner is unlinked.
*/
static int
zfs_xattr_owner_unlinked(znode_t *zp)
{
int unlinked = 0;
znode_t *dzp;
igrab(ZTOI(zp));
/*
* if zp is XATTR node, keep walking up via z_xattr_parent until we
* get the owner
*/
while (zp->z_pflags & ZFS_XATTR) {
ASSERT3U(zp->z_xattr_parent, !=, 0);
if (zfs_zget(ZTOZSB(zp), zp->z_xattr_parent, &dzp) != 0) {
unlinked = 1;
break;
}
iput(ZTOI(zp));
zp = dzp;
unlinked = zp->z_unlinked;
}
iput(ZTOI(zp));
return (unlinked);
}
static int
__zpl_xattr_get(struct inode *ip, const char *name, void *value, size_t size,
cred_t *cr)
{
znode_t *zp = ITOZ(ip);
zfs_sb_t *zsb = ZTOZSB(zp);
int error;
ASSERT(RW_LOCK_HELD(&zp->z_xattr_lock));
if (zsb->z_use_sa && zp->z_is_sa) {
error = zpl_xattr_get_sa(ip, name, value, size);
if (error != -ENOENT)
goto out;
}
error = zpl_xattr_get_dir(ip, name, value, size, cr);
out:
if (error == -ENOENT)
error = -ENODATA;
return (error);
}
/*
* Decide whether it is okay to remove within a sticky directory.
*
* In sticky directories, write access is not sufficient;
* you can remove entries from a directory only if:
*
* you own the directory,
* you own the entry,
* the entry is a plain file and you have write access,
* or you are privileged (checked in secpolicy...).
*
* The function returns 0 if remove access is granted.
*/
int
zfs_sticky_remove_access(znode_t *zdp, znode_t *zp, cred_t *cr)
{
uid_t uid;
uid_t downer;
uid_t fowner;
zfs_sb_t *zsb = ZTOZSB(zdp);
if (zsb->z_replay)
return (0);
if ((zdp->z_mode & S_ISVTX) == 0)
return (0);
downer = zfs_fuid_map_id(zsb, zdp->z_uid, cr, ZFS_OWNER);
fowner = zfs_fuid_map_id(zsb, zp->z_uid, cr, ZFS_OWNER);
if ((uid = crgetuid(cr)) == downer || uid == fowner ||
(S_ISDIR(ZTOI(zp)->i_mode) &&
zfs_zaccess(zp, ACE_WRITE_DATA, 0, B_FALSE, cr) == 0))
return (0);
else
return (secpolicy_vnode_remove(cr));
}
static int
zfs_dropname(zfs_dirlock_t *dl, znode_t *zp, znode_t *dzp, dmu_tx_t *tx,
int flag)
{
int error;
if (ZTOZSB(zp)->z_norm) {
if (((ZTOZSB(zp)->z_case == ZFS_CASE_INSENSITIVE) &&
(flag & ZCIEXACT)) ||
((ZTOZSB(zp)->z_case == ZFS_CASE_MIXED) &&
!(flag & ZCILOOK)))
error = zap_remove_norm(ZTOZSB(zp)->z_os,
dzp->z_id, dl->dl_name, MT_EXACT, tx);
else
error = zap_remove_norm(ZTOZSB(zp)->z_os,
dzp->z_id, dl->dl_name, MT_FIRST, tx);
} else {
error = zap_remove(ZTOZSB(zp)->z_os,
dzp->z_id, dl->dl_name, tx);
}
return (error);
}
void
zfs_sa_upgrade(sa_handle_t *hdl, dmu_tx_t *tx)
{
dmu_buf_t *db = sa_get_db(hdl);
znode_t *zp = sa_get_userdata(hdl);
zfs_sb_t *zsb = ZTOZSB(zp);
int count = 0;
sa_bulk_attr_t *bulk, *sa_attrs;
zfs_acl_locator_cb_t locate = { 0 };
uint64_t uid, gid, mode, rdev, xattr, parent;
uint64_t crtime[2], mtime[2], ctime[2];
zfs_acl_phys_t znode_acl;
char scanstamp[AV_SCANSTAMP_SZ];
boolean_t drop_lock = B_FALSE;
/*
* No upgrade if ACL isn't cached
* since we won't know which locks are held
* and ready the ACL would require special "locked"
* interfaces that would be messy
*/
if (zp->z_acl_cached == NULL || S_ISLNK(ZTOI(zp)->i_mode))
return;
/*
* If the z_lock is held and we aren't the owner
* the just return since we don't want to deadlock
* trying to update the status of z_is_sa. This
* file can then be upgraded at a later time.
*
* Otherwise, we know we are doing the
* sa_update() that caused us to enter this function.
*/
if (mutex_owner(&zp->z_lock) != curthread) {
if (mutex_tryenter(&zp->z_lock) == 0)
return;
else
drop_lock = B_TRUE;
}
/* First do a bulk query of the attributes that aren't cached */
bulk = kmem_alloc(sizeof (sa_bulk_attr_t) * 20, KM_SLEEP);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zsb), NULL, &mtime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zsb), NULL, &ctime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CRTIME(zsb), NULL, &crtime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MODE(zsb), NULL, &mode, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_PARENT(zsb), NULL, &parent, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_XATTR(zsb), NULL, &xattr, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_RDEV(zsb), NULL, &rdev, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_UID(zsb), NULL, &uid, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GID(zsb), NULL, &gid, 8);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_ZNODE_ACL(zsb), NULL,
&znode_acl, 88);
if (sa_bulk_lookup_locked(hdl, bulk, count) != 0) {
kmem_free(bulk, sizeof (sa_bulk_attr_t) * 20);
goto done;
}
/*
* While the order here doesn't matter its best to try and organize
* it is such a way to pick up an already existing layout number
*/
count = 0;
sa_attrs = kmem_zalloc(sizeof (sa_bulk_attr_t) * 20, KM_SLEEP);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_MODE(zsb), NULL, &mode, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_SIZE(zsb), NULL,
&zp->z_size, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_GEN(zsb),
NULL, &zp->z_gen, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_UID(zsb), NULL, &uid, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_GID(zsb), NULL, &gid, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_PARENT(zsb),
NULL, &parent, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_FLAGS(zsb), NULL,
&zp->z_pflags, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_ATIME(zsb), NULL,
zp->z_atime, 16);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_MTIME(zsb), NULL,
&mtime, 16);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_CTIME(zsb), NULL,
&ctime, 16);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_CRTIME(zsb), NULL,
&crtime, 16);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_LINKS(zsb), NULL,
&zp->z_links, 8);
if (S_ISBLK(ZTOI(zp)->i_mode) || S_ISCHR(ZTOI(zp)->i_mode))
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_RDEV(zsb), NULL,
&rdev, 8);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_DACL_COUNT(zsb), NULL,
&zp->z_acl_cached->z_acl_count, 8);
if (zp->z_acl_cached->z_version < ZFS_ACL_VERSION_FUID)
zfs_acl_xform(zp, zp->z_acl_cached, CRED());
locate.cb_aclp = zp->z_acl_cached;
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_DACL_ACES(zsb),
zfs_acl_data_locator, &locate, zp->z_acl_cached->z_acl_bytes);
if (xattr)
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_XATTR(zsb),
NULL, &xattr, 8);
/* if scanstamp then add scanstamp */
if (zp->z_pflags & ZFS_BONUS_SCANSTAMP) {
bcopy((caddr_t)db->db_data + ZFS_OLD_ZNODE_PHYS_SIZE,
scanstamp, AV_SCANSTAMP_SZ);
SA_ADD_BULK_ATTR(sa_attrs, count, SA_ZPL_SCANSTAMP(zsb),
NULL, scanstamp, AV_SCANSTAMP_SZ);
zp->z_pflags &= ~ZFS_BONUS_SCANSTAMP;
}
VERIFY(dmu_set_bonustype(db, DMU_OT_SA, tx) == 0);
VERIFY(sa_replace_all_by_template_locked(hdl, sa_attrs,
count, tx) == 0);
if (znode_acl.z_acl_extern_obj)
VERIFY(0 == dmu_object_free(zsb->z_os,
znode_acl.z_acl_extern_obj, tx));
zp->z_is_sa = B_TRUE;
kmem_free(sa_attrs, sizeof (sa_bulk_attr_t) * 20);
kmem_free(bulk, sizeof (sa_bulk_attr_t) * 20);
done:
if (drop_lock)
mutex_exit(&zp->z_lock);
}
int
zfs_make_xattrdir(znode_t *zp, vattr_t *vap, struct inode **xipp, cred_t *cr)
{
zfsvfs_t *zfsvfs = ZTOZSB(zp);
znode_t *xzp;
dmu_tx_t *tx;
int error;
zfs_acl_ids_t acl_ids;
boolean_t fuid_dirtied;
#ifdef DEBUG
uint64_t parent;
#endif
*xipp = NULL;
if ((error = zfs_zaccess(zp, ACE_WRITE_NAMED_ATTRS, 0, B_FALSE, cr)))
return (error);
if ((error = zfs_acl_ids_create(zp, IS_XATTR, vap, cr, NULL,
&acl_ids)) != 0)
return (error);
if (zfs_acl_ids_overquota(zfsvfs, &acl_ids)) {
zfs_acl_ids_free(&acl_ids);
return (SET_ERROR(EDQUOT));
}
tx = dmu_tx_create(zfsvfs->z_os);
dmu_tx_hold_sa_create(tx, acl_ids.z_aclp->z_acl_bytes +
ZFS_SA_BASE_ATTR_SIZE);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_TRUE);
dmu_tx_hold_zap(tx, DMU_NEW_OBJECT, FALSE, NULL);
fuid_dirtied = zfsvfs->z_fuid_dirty;
if (fuid_dirtied)
zfs_fuid_txhold(zfsvfs, tx);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
zfs_acl_ids_free(&acl_ids);
dmu_tx_abort(tx);
return (error);
}
zfs_mknode(zp, vap, tx, cr, IS_XATTR, &xzp, &acl_ids);
if (fuid_dirtied)
zfs_fuid_sync(zfsvfs, tx);
#ifdef DEBUG
error = sa_lookup(xzp->z_sa_hdl, SA_ZPL_PARENT(zfsvfs),
&parent, sizeof (parent));
ASSERT(error == 0 && parent == zp->z_id);
#endif
VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_XATTR(zfsvfs), &xzp->z_id,
sizeof (xzp->z_id), tx));
if (!zp->z_unlinked)
(void) zfs_log_create(zfsvfs->z_log, tx, TX_MKXATTR, zp,
xzp, "", NULL, acl_ids.z_fuidp, vap);
zfs_acl_ids_free(&acl_ids);
dmu_tx_commit(tx);
*xipp = ZTOI(xzp);
return (0);
}
/*
* Unlink zp from dl, and mark zp for deletion if this was the last link. Can
* fail if zp is a mount point (EBUSY) or a non-empty directory (ENOTEMPTY).
* If 'unlinkedp' is NULL, we put unlinked znodes on the unlinked list.
* If it's non-NULL, we use it to indicate whether the znode needs deletion,
* and it's the caller's job to do it.
*/
int
zfs_link_destroy(zfs_dirlock_t *dl, znode_t *zp, dmu_tx_t *tx, int flag,
boolean_t *unlinkedp)
{
znode_t *dzp = dl->dl_dzp;
zfsvfs_t *zfsvfs = ZTOZSB(dzp);
int zp_is_dir = S_ISDIR(ZTOI(zp)->i_mode);
boolean_t unlinked = B_FALSE;
sa_bulk_attr_t bulk[5];
uint64_t mtime[2], ctime[2];
uint64_t links;
int count = 0;
int error;
#ifdef HAVE_DNLC
dnlc_remove(ZTOI(dzp), dl->dl_name);
#endif /* HAVE_DNLC */
if (!(flag & ZRENAMING)) {
mutex_enter(&zp->z_lock);
if (zp_is_dir && !zfs_dirempty(zp)) {
mutex_exit(&zp->z_lock);
return (SET_ERROR(ENOTEMPTY));
}
/*
* If we get here, we are going to try to remove the object.
* First try removing the name from the directory; if that
* fails, return the error.
*/
error = zfs_dropname(dl, zp, dzp, tx, flag);
if (error != 0) {
mutex_exit(&zp->z_lock);
return (error);
}
if (ZTOI(zp)->i_nlink <= zp_is_dir) {
zfs_panic_recover("zfs: link count on %lu is %u, "
"should be at least %u", zp->z_id,
(int)ZTOI(zp)->i_nlink, zp_is_dir + 1);
set_nlink(ZTOI(zp), zp_is_dir + 1);
}
drop_nlink(ZTOI(zp));
if (ZTOI(zp)->i_nlink == zp_is_dir) {
zp->z_unlinked = B_TRUE;
clear_nlink(ZTOI(zp));
unlinked = B_TRUE;
} else {
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs),
NULL, &ctime, sizeof (ctime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs),
NULL, &zp->z_pflags, sizeof (zp->z_pflags));
zfs_tstamp_update_setup(zp, STATE_CHANGED, mtime,
ctime);
}
links = ZTOI(zp)->i_nlink;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs),
NULL, &links, sizeof (links));
error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
count = 0;
ASSERT(error == 0);
mutex_exit(&zp->z_lock);
} else {
error = zfs_dropname(dl, zp, dzp, tx, flag);
if (error != 0)
return (error);
}
mutex_enter(&dzp->z_lock);
dzp->z_size--; /* one dirent removed */
if (zp_is_dir)
drop_nlink(ZTOI(dzp)); /* ".." link from zp */
links = ZTOI(dzp)->i_nlink;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs),
NULL, &links, sizeof (links));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zfsvfs),
NULL, &dzp->z_size, sizeof (dzp->z_size));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs),
NULL, ctime, sizeof (ctime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zfsvfs),
NULL, mtime, sizeof (mtime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs),
NULL, &dzp->z_pflags, sizeof (dzp->z_pflags));
zfs_tstamp_update_setup(dzp, CONTENT_MODIFIED, mtime, ctime);
error = sa_bulk_update(dzp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
mutex_exit(&dzp->z_lock);
if (unlinkedp != NULL)
*unlinkedp = unlinked;
else if (unlinked)
zfs_unlinked_add(zp, tx);
return (0);
}
/*
* Link zp into dl. Can only fail if zp has been unlinked.
*/
int
zfs_link_create(zfs_dirlock_t *dl, znode_t *zp, dmu_tx_t *tx, int flag)
{
znode_t *dzp = dl->dl_dzp;
zfsvfs_t *zfsvfs = ZTOZSB(zp);
uint64_t value;
int zp_is_dir = S_ISDIR(ZTOI(zp)->i_mode);
sa_bulk_attr_t bulk[5];
uint64_t mtime[2], ctime[2];
uint64_t links;
int count = 0;
int error;
mutex_enter(&zp->z_lock);
if (!(flag & ZRENAMING)) {
if (zp->z_unlinked) { /* no new links to unlinked zp */
ASSERT(!(flag & (ZNEW | ZEXISTS)));
mutex_exit(&zp->z_lock);
return (SET_ERROR(ENOENT));
}
if (!(flag & ZNEW)) {
/*
* ZNEW nodes come from zfs_mknode() where the link
* count has already been initialised
*/
inc_nlink(ZTOI(zp));
links = ZTOI(zp)->i_nlink;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs),
NULL, &links, sizeof (links));
}
}
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_PARENT(zfsvfs), NULL,
&dzp->z_id, sizeof (dzp->z_id));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs), NULL,
&zp->z_pflags, sizeof (zp->z_pflags));
if (!(flag & ZNEW)) {
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs), NULL,
ctime, sizeof (ctime));
zfs_tstamp_update_setup(zp, STATE_CHANGED, mtime,
ctime);
}
error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
mutex_exit(&zp->z_lock);
mutex_enter(&dzp->z_lock);
dzp->z_size++;
if (zp_is_dir)
inc_nlink(ZTOI(dzp));
links = ZTOI(dzp)->i_nlink;
count = 0;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zfsvfs), NULL,
&dzp->z_size, sizeof (dzp->z_size));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zfsvfs), NULL,
&links, sizeof (links));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zfsvfs), NULL,
mtime, sizeof (mtime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zfsvfs), NULL,
ctime, sizeof (ctime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zfsvfs), NULL,
&dzp->z_pflags, sizeof (dzp->z_pflags));
zfs_tstamp_update_setup(dzp, CONTENT_MODIFIED, mtime, ctime);
error = sa_bulk_update(dzp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
mutex_exit(&dzp->z_lock);
value = zfs_dirent(zp, zp->z_mode);
error = zap_add(ZTOZSB(zp)->z_os, dzp->z_id, dl->dl_name,
8, 1, &value, tx);
ASSERT(error == 0);
return (0);
}
void
zfs_rmnode(znode_t *zp)
{
zfsvfs_t *zfsvfs = ZTOZSB(zp);
objset_t *os = zfsvfs->z_os;
znode_t *xzp = NULL;
dmu_tx_t *tx;
uint64_t acl_obj;
uint64_t xattr_obj;
uint64_t links;
int error;
ASSERT(ZTOI(zp)->i_nlink == 0);
ASSERT(atomic_read(&ZTOI(zp)->i_count) == 0);
/*
* If this is an attribute directory, purge its contents.
*/
if (S_ISDIR(ZTOI(zp)->i_mode) && (zp->z_pflags & ZFS_XATTR)) {
if (zfs_purgedir(zp) != 0) {
/*
* Not enough space to delete some xattrs.
* Leave it in the unlinked set.
*/
zfs_znode_dmu_fini(zp);
return;
}
}
/*
* Free up all the data in the file. We don't do this for directories
* because we need truncate and remove to be in the same tx, like in
* zfs_znode_delete(). Otherwise, if we crash here we'll end up with
* an inconsistent truncated zap object in the delete queue. Note a
* truncated file is harmless since it only contains user data.
*/
if (S_ISREG(ZTOI(zp)->i_mode)) {
error = dmu_free_long_range(os, zp->z_id, 0, DMU_OBJECT_END);
if (error) {
/*
* Not enough space or we were interrupted by unmount.
* Leave the file in the unlinked set.
*/
zfs_znode_dmu_fini(zp);
return;
}
}
/*
* If the file has extended attributes, we're going to unlink
* the xattr dir.
*/
error = sa_lookup(zp->z_sa_hdl, SA_ZPL_XATTR(zfsvfs),
&xattr_obj, sizeof (xattr_obj));
if (error == 0 && xattr_obj) {
error = zfs_zget(zfsvfs, xattr_obj, &xzp);
ASSERT(error == 0);
}
acl_obj = zfs_external_acl(zp);
/*
* Set up the final transaction.
*/
tx = dmu_tx_create(os);
dmu_tx_hold_free(tx, zp->z_id, 0, DMU_OBJECT_END);
dmu_tx_hold_zap(tx, zfsvfs->z_unlinkedobj, FALSE, NULL);
if (xzp) {
dmu_tx_hold_zap(tx, zfsvfs->z_unlinkedobj, TRUE, NULL);
dmu_tx_hold_sa(tx, xzp->z_sa_hdl, B_FALSE);
}
if (acl_obj)
dmu_tx_hold_free(tx, acl_obj, 0, DMU_OBJECT_END);
zfs_sa_upgrade_txholds(tx, zp);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
/*
* Not enough space to delete the file. Leave it in the
* unlinked set, leaking it until the fs is remounted (at
* which point we'll call zfs_unlinked_drain() to process it).
*/
dmu_tx_abort(tx);
zfs_znode_dmu_fini(zp);
goto out;
}
if (xzp) {
ASSERT(error == 0);
mutex_enter(&xzp->z_lock);
xzp->z_unlinked = B_TRUE; /* mark xzp for deletion */
clear_nlink(ZTOI(xzp)); /* no more links to it */
links = 0;
VERIFY(0 == sa_update(xzp->z_sa_hdl, SA_ZPL_LINKS(zfsvfs),
&links, sizeof (links), tx));
mutex_exit(&xzp->z_lock);
zfs_unlinked_add(xzp, tx);
}
/* Remove this znode from the unlinked set */
VERIFY3U(0, ==,
zap_remove_int(zfsvfs->z_os, zfsvfs->z_unlinkedobj, zp->z_id, tx));
zfs_znode_delete(zp, tx);
dmu_tx_commit(tx);
out:
if (xzp)
zfs_iput_async(ZTOI(xzp));
}
/*
* Lock a directory entry. A dirlock on <dzp, name> protects that name
* in dzp's directory zap object. As long as you hold a dirlock, you can
* assume two things: (1) dzp cannot be reaped, and (2) no other thread
* can change the zap entry for (i.e. link or unlink) this name.
*
* Input arguments:
* dzp - znode for directory
* name - name of entry to lock
* flag - ZNEW: if the entry already exists, fail with EEXIST.
* ZEXISTS: if the entry does not exist, fail with ENOENT.
* ZSHARED: allow concurrent access with other ZSHARED callers.
* ZXATTR: we want dzp's xattr directory
* ZCILOOK: On a mixed sensitivity file system,
* this lookup should be case-insensitive.
* ZCIEXACT: On a purely case-insensitive file system,
* this lookup should be case-sensitive.
* ZRENAMING: we are locking for renaming, force narrow locks
* ZHAVELOCK: Don't grab the z_name_lock for this call. The
* current thread already holds it.
*
* Output arguments:
* zpp - pointer to the znode for the entry (NULL if there isn't one)
* dlpp - pointer to the dirlock for this entry (NULL on error)
* direntflags - (case-insensitive lookup only)
* flags if multiple case-sensitive matches exist in directory
* realpnp - (case-insensitive lookup only)
* actual name matched within the directory
*
* Return value: 0 on success or errno on failure.
*
* NOTE: Always checks for, and rejects, '.' and '..'.
* NOTE: For case-insensitive file systems we take wide locks (see below),
* but return znode pointers to a single match.
*/
int
zfs_dirent_lock(zfs_dirlock_t **dlpp, znode_t *dzp, char *name, znode_t **zpp,
int flag, int *direntflags, pathname_t *realpnp)
{
zfsvfs_t *zfsvfs = ZTOZSB(dzp);
zfs_dirlock_t *dl;
boolean_t update;
matchtype_t mt = 0;
uint64_t zoid;
#ifdef HAVE_DNLC
vnode_t *vp = NULL;
#endif /* HAVE_DNLC */
int error = 0;
int cmpflags;
*zpp = NULL;
*dlpp = NULL;
/*
* Verify that we are not trying to lock '.', '..', or '.zfs'
*/
if ((name[0] == '.' &&
(name[1] == '\0' || (name[1] == '.' && name[2] == '\0'))) ||
(zfs_has_ctldir(dzp) && strcmp(name, ZFS_CTLDIR_NAME) == 0))
return (SET_ERROR(EEXIST));
/*
* Case sensitivity and normalization preferences are set when
* the file system is created. These are stored in the
* zfsvfs->z_case and zfsvfs->z_norm fields. These choices
* affect what vnodes can be cached in the DNLC, how we
* perform zap lookups, and the "width" of our dirlocks.
*
* A normal dirlock locks a single name. Note that with
* normalization a name can be composed multiple ways, but
* when normalized, these names all compare equal. A wide
* dirlock locks multiple names. We need these when the file
* system is supporting mixed-mode access. It is sometimes
* necessary to lock all case permutations of file name at
* once so that simultaneous case-insensitive/case-sensitive
* behaves as rationally as possible.
*/
/*
* When matching we may need to normalize & change case according to
* FS settings.
*
* Note that a normalized match is necessary for a case insensitive
* filesystem when the lookup request is not exact because normalization
* can fold case independent of normalizing code point sequences.
*
* See the table above zfs_dropname().
*/
if (zfsvfs->z_norm != 0) {
mt = MT_NORMALIZE;
/*
* Determine if the match needs to honor the case specified in
* lookup, and if so keep track of that so that during
* normalization we don't fold case.
*/
if ((zfsvfs->z_case == ZFS_CASE_INSENSITIVE &&
(flag & ZCIEXACT)) ||
(zfsvfs->z_case == ZFS_CASE_MIXED && !(flag & ZCILOOK))) {
mt |= MT_MATCH_CASE;
}
}
/*
* Only look in or update the DNLC if we are looking for the
* name on a file system that does not require normalization
* or case folding. We can also look there if we happen to be
* on a non-normalizing, mixed sensitivity file system IF we
* are looking for the exact name.
*
* Maybe can add TO-UPPERed version of name to dnlc in ci-only
* case for performance improvement?
*/
update = !zfsvfs->z_norm ||
(zfsvfs->z_case == ZFS_CASE_MIXED &&
!(zfsvfs->z_norm & ~U8_TEXTPREP_TOUPPER) && !(flag & ZCILOOK));
/*
* ZRENAMING indicates we are in a situation where we should
* take narrow locks regardless of the file system's
* preferences for normalizing and case folding. This will
* prevent us deadlocking trying to grab the same wide lock
* twice if the two names happen to be case-insensitive
* matches.
*/
if (flag & ZRENAMING)
cmpflags = 0;
else
cmpflags = zfsvfs->z_norm;
/*
* Wait until there are no locks on this name.
*
* Don't grab the the lock if it is already held. However, cannot
* have both ZSHARED and ZHAVELOCK together.
*/
ASSERT(!(flag & ZSHARED) || !(flag & ZHAVELOCK));
if (!(flag & ZHAVELOCK))
rw_enter(&dzp->z_name_lock, RW_READER);
mutex_enter(&dzp->z_lock);
for (;;) {
if (dzp->z_unlinked && !(flag & ZXATTR)) {
mutex_exit(&dzp->z_lock);
if (!(flag & ZHAVELOCK))
rw_exit(&dzp->z_name_lock);
return (SET_ERROR(ENOENT));
}
for (dl = dzp->z_dirlocks; dl != NULL; dl = dl->dl_next) {
if ((u8_strcmp(name, dl->dl_name, 0, cmpflags,
U8_UNICODE_LATEST, &error) == 0) || error != 0)
break;
}
if (error != 0) {
mutex_exit(&dzp->z_lock);
if (!(flag & ZHAVELOCK))
rw_exit(&dzp->z_name_lock);
return (SET_ERROR(ENOENT));
}
if (dl == NULL) {
/*
* Allocate a new dirlock and add it to the list.
*/
dl = kmem_alloc(sizeof (zfs_dirlock_t), KM_SLEEP);
cv_init(&dl->dl_cv, NULL, CV_DEFAULT, NULL);
dl->dl_name = name;
dl->dl_sharecnt = 0;
dl->dl_namelock = 0;
dl->dl_namesize = 0;
dl->dl_dzp = dzp;
dl->dl_next = dzp->z_dirlocks;
dzp->z_dirlocks = dl;
break;
}
if ((flag & ZSHARED) && dl->dl_sharecnt != 0)
break;
cv_wait(&dl->dl_cv, &dzp->z_lock);
}
/*
* If the z_name_lock was NOT held for this dirlock record it.
*/
if (flag & ZHAVELOCK)
dl->dl_namelock = 1;
if ((flag & ZSHARED) && ++dl->dl_sharecnt > 1 && dl->dl_namesize == 0) {
/*
* We're the second shared reference to dl. Make a copy of
* dl_name in case the first thread goes away before we do.
* Note that we initialize the new name before storing its
* pointer into dl_name, because the first thread may load
* dl->dl_name at any time. It'll either see the old value,
* which belongs to it, or the new shared copy; either is OK.
*/
dl->dl_namesize = strlen(dl->dl_name) + 1;
name = kmem_alloc(dl->dl_namesize, KM_SLEEP);
bcopy(dl->dl_name, name, dl->dl_namesize);
dl->dl_name = name;
}
mutex_exit(&dzp->z_lock);
/*
* We have a dirlock on the name. (Note that it is the dirlock,
* not the dzp's z_lock, that protects the name in the zap object.)
* See if there's an object by this name; if so, put a hold on it.
*/
if (flag & ZXATTR) {
error = sa_lookup(dzp->z_sa_hdl, SA_ZPL_XATTR(zfsvfs), &zoid,
sizeof (zoid));
if (error == 0)
error = (zoid == 0 ? SET_ERROR(ENOENT) : 0);
} else {
#ifdef HAVE_DNLC
if (update)
vp = dnlc_lookup(ZTOI(dzp), name);
if (vp == DNLC_NO_VNODE) {
iput(vp);
error = SET_ERROR(ENOENT);
} else if (vp) {
if (flag & ZNEW) {
zfs_dirent_unlock(dl);
iput(vp);
return (SET_ERROR(EEXIST));
}
*dlpp = dl;
*zpp = VTOZ(vp);
return (0);
} else {
error = zfs_match_find(zfsvfs, dzp, name, mt,
update, direntflags, realpnp, &zoid);
}
#else
error = zfs_match_find(zfsvfs, dzp, name, mt,
update, direntflags, realpnp, &zoid);
#endif /* HAVE_DNLC */
}
if (error) {
if (error != ENOENT || (flag & ZEXISTS)) {
zfs_dirent_unlock(dl);
return (error);
}
} else {
if (flag & ZNEW) {
zfs_dirent_unlock(dl);
return (SET_ERROR(EEXIST));
}
error = zfs_zget(zfsvfs, zoid, zpp);
if (error) {
zfs_dirent_unlock(dl);
return (error);
}
#ifdef HAVE_DNLC
if (!(flag & ZXATTR) && update)
dnlc_update(ZTOI(dzp), name, ZTOI(*zpp));
#endif /* HAVE_DNLC */
}
*dlpp = dl;
return (0);
}
/*
* Increase the file length
*
* IN: zp - znode of file to free data in.
* end - new end-of-file
*
* RETURN: 0 on success, error code on failure
*/
static int
zfs_extend(znode_t *zp, uint64_t end)
{
zfs_sb_t *zsb = ZTOZSB(zp);
dmu_tx_t *tx;
rl_t *rl;
uint64_t newblksz;
int error;
/*
* We will change zp_size, lock the whole file.
*/
rl = zfs_range_lock(zp, 0, UINT64_MAX, RL_WRITER);
/*
* Nothing to do if file already at desired length.
*/
if (end <= zp->z_size) {
zfs_range_unlock(rl);
return (0);
}
tx = dmu_tx_create(zsb->z_os);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
if (end > zp->z_blksz &&
(!ISP2(zp->z_blksz) || zp->z_blksz < zsb->z_max_blksz)) {
/*
* We are growing the file past the current block size.
*/
if (zp->z_blksz > ZTOZSB(zp)->z_max_blksz) {
ASSERT(!ISP2(zp->z_blksz));
newblksz = MIN(end, SPA_MAXBLOCKSIZE);
} else {
newblksz = MIN(end, ZTOZSB(zp)->z_max_blksz);
}
dmu_tx_hold_write(tx, zp->z_id, 0, newblksz);
} else {
newblksz = 0;
}
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
dmu_tx_abort(tx);
zfs_range_unlock(rl);
return (error);
}
if (newblksz)
zfs_grow_blocksize(zp, newblksz, tx);
zp->z_size = end;
VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_SIZE(ZTOZSB(zp)),
&zp->z_size, sizeof (zp->z_size), tx));
zfs_range_unlock(rl);
dmu_tx_commit(tx);
return (0);
}
/*
* Truncate a file
*
* IN: zp - znode of file to free data in.
* end - new end-of-file.
*
* RETURN: 0 on success, error code on failure
*/
static int
zfs_trunc(znode_t *zp, uint64_t end)
{
zfs_sb_t *zsb = ZTOZSB(zp);
dmu_tx_t *tx;
rl_t *rl;
int error;
sa_bulk_attr_t bulk[2];
int count = 0;
/*
* We will change zp_size, lock the whole file.
*/
rl = zfs_range_lock(zp, 0, UINT64_MAX, RL_WRITER);
/*
* Nothing to do if file already at desired length.
*/
if (end >= zp->z_size) {
zfs_range_unlock(rl);
return (0);
}
error = dmu_free_long_range(zsb->z_os, zp->z_id, end, -1);
if (error) {
zfs_range_unlock(rl);
return (error);
}
top:
tx = dmu_tx_create(zsb->z_os);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
error = dmu_tx_assign(tx, TXG_NOWAIT);
if (error) {
if (error == ERESTART) {
dmu_tx_wait(tx);
dmu_tx_abort(tx);
goto top;
}
dmu_tx_abort(tx);
zfs_range_unlock(rl);
return (error);
}
zp->z_size = end;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zsb),
NULL, &zp->z_size, sizeof (zp->z_size));
if (end == 0) {
zp->z_pflags &= ~ZFS_SPARSE;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zsb),
NULL, &zp->z_pflags, 8);
}
VERIFY(sa_bulk_update(zp->z_sa_hdl, bulk, count, tx) == 0);
dmu_tx_commit(tx);
zfs_range_unlock(rl);
return (0);
}
/*
* Handles TX_CREATE, TX_CREATE_ATTR, TX_MKDIR, TX_MKDIR_ATTR and
* TK_MKXATTR transactions.
*
* TX_CREATE and TX_MKDIR are standard creates, but they may have FUID
* domain information appended prior to the name. In this case the
* uid/gid in the log record will be a log centric FUID.
*
* TX_CREATE_ACL_ATTR and TX_MKDIR_ACL_ATTR handle special creates that
* may contain attributes, ACL and optional fuid information.
*
* TX_CREATE_ACL and TX_MKDIR_ACL handle special creates that specify
* and ACL and normal users/groups in the ACEs.
*
* There may be an optional xvattr attribute information similar
* to zfs_log_setattr.
*
* Also, after the file name "domain" strings may be appended.
*/
void
zfs_log_create(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
znode_t *dzp, znode_t *zp, char *name, vsecattr_t *vsecp,
zfs_fuid_info_t *fuidp, vattr_t *vap)
{
itx_t *itx;
lr_create_t *lr;
lr_acl_create_t *lracl;
size_t aclsize = 0;
size_t xvatsize = 0;
size_t txsize;
xvattr_t *xvap = (xvattr_t *)vap;
void *end;
size_t lrsize;
size_t namesize = strlen(name) + 1;
size_t fuidsz = 0;
if (zil_replaying(zilog, tx))
return;
/*
* If we have FUIDs present then add in space for
* domains and ACE fuid's if any.
*/
if (fuidp) {
fuidsz += fuidp->z_domain_str_sz;
fuidsz += fuidp->z_fuid_cnt * sizeof (uint64_t);
}
if (vap->va_mask & ATTR_XVATTR)
xvatsize = ZIL_XVAT_SIZE(xvap->xva_mapsize);
if ((int)txtype == TX_CREATE_ATTR || (int)txtype == TX_MKDIR_ATTR ||
(int)txtype == TX_CREATE || (int)txtype == TX_MKDIR ||
(int)txtype == TX_MKXATTR) {
txsize = sizeof (*lr) + namesize + fuidsz + xvatsize;
lrsize = sizeof (*lr);
} else {
txsize =
sizeof (lr_acl_create_t) + namesize + fuidsz +
ZIL_ACE_LENGTH(aclsize) + xvatsize;
lrsize = sizeof (lr_acl_create_t);
}
itx = zil_itx_create(txtype, txsize);
lr = (lr_create_t *)&itx->itx_lr;
lr->lr_doid = dzp->z_id;
lr->lr_foid = zp->z_id;
lr->lr_mode = zp->z_mode;
if (!IS_EPHEMERAL(zp->z_uid)) {
lr->lr_uid = (uint64_t)zp->z_uid;
} else {
lr->lr_uid = fuidp->z_fuid_owner;
}
if (!IS_EPHEMERAL(zp->z_gid)) {
lr->lr_gid = (uint64_t)zp->z_gid;
} else {
lr->lr_gid = fuidp->z_fuid_group;
}
(void) sa_lookup(zp->z_sa_hdl, SA_ZPL_GEN(ZTOZSB(zp)), &lr->lr_gen,
sizeof (uint64_t));
(void) sa_lookup(zp->z_sa_hdl, SA_ZPL_CRTIME(ZTOZSB(zp)),
lr->lr_crtime, sizeof (uint64_t) * 2);
if (sa_lookup(zp->z_sa_hdl, SA_ZPL_RDEV(ZTOZSB(zp)), &lr->lr_rdev,
sizeof (lr->lr_rdev)) != 0)
lr->lr_rdev = 0;
/*
* Fill in xvattr info if any
*/
if (vap->va_mask & ATTR_XVATTR) {
zfs_log_xvattr((lr_attr_t *)((caddr_t)lr + lrsize), xvap);
end = (caddr_t)lr + lrsize + xvatsize;
} else {
end = (caddr_t)lr + lrsize;
}
/* Now fill in any ACL info */
if (vsecp) {
lracl = (lr_acl_create_t *)&itx->itx_lr;
lracl->lr_aclcnt = vsecp->vsa_aclcnt;
lracl->lr_acl_bytes = aclsize;
lracl->lr_domcnt = fuidp ? fuidp->z_domain_cnt : 0;
lracl->lr_fuidcnt = fuidp ? fuidp->z_fuid_cnt : 0;
if (vsecp->vsa_aclflags & VSA_ACE_ACLFLAGS)
lracl->lr_acl_flags = (uint64_t)vsecp->vsa_aclflags;
else
lracl->lr_acl_flags = 0;
bcopy(vsecp->vsa_aclentp, end, aclsize);
end = (caddr_t)end + ZIL_ACE_LENGTH(aclsize);
}
/* drop in FUID info */
if (fuidp) {
end = zfs_log_fuid_ids(fuidp, end);
end = zfs_log_fuid_domains(fuidp, end);
}
/*
* Now place file name in log record
*/
bcopy(name, end, namesize);
zil_itx_assign(zilog, itx, tx);
}
/*
* Update in-core attributes. It is assumed the caller will be doing an
* sa_bulk_update to push the changes out.
*/
void
zfs_xvattr_set(znode_t *zp, xvattr_t *xvap, dmu_tx_t *tx)
{
xoptattr_t *xoap;
xoap = xva_getxoptattr(xvap);
ASSERT(xoap);
if (XVA_ISSET_REQ(xvap, XAT_CREATETIME)) {
uint64_t times[2];
ZFS_TIME_ENCODE(&xoap->xoa_createtime, times);
(void) sa_update(zp->z_sa_hdl, SA_ZPL_CRTIME(ZTOZSB(zp)),
×, sizeof (times), tx);
XVA_SET_RTN(xvap, XAT_CREATETIME);
}
if (XVA_ISSET_REQ(xvap, XAT_READONLY)) {
ZFS_ATTR_SET(zp, ZFS_READONLY, xoap->xoa_readonly,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_READONLY);
}
if (XVA_ISSET_REQ(xvap, XAT_HIDDEN)) {
ZFS_ATTR_SET(zp, ZFS_HIDDEN, xoap->xoa_hidden,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_HIDDEN);
}
if (XVA_ISSET_REQ(xvap, XAT_SYSTEM)) {
ZFS_ATTR_SET(zp, ZFS_SYSTEM, xoap->xoa_system,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_SYSTEM);
}
if (XVA_ISSET_REQ(xvap, XAT_ARCHIVE)) {
ZFS_ATTR_SET(zp, ZFS_ARCHIVE, xoap->xoa_archive,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_ARCHIVE);
}
if (XVA_ISSET_REQ(xvap, XAT_IMMUTABLE)) {
ZFS_ATTR_SET(zp, ZFS_IMMUTABLE, xoap->xoa_immutable,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_IMMUTABLE);
}
if (XVA_ISSET_REQ(xvap, XAT_NOUNLINK)) {
ZFS_ATTR_SET(zp, ZFS_NOUNLINK, xoap->xoa_nounlink,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_NOUNLINK);
}
if (XVA_ISSET_REQ(xvap, XAT_APPENDONLY)) {
ZFS_ATTR_SET(zp, ZFS_APPENDONLY, xoap->xoa_appendonly,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_APPENDONLY);
}
if (XVA_ISSET_REQ(xvap, XAT_NODUMP)) {
ZFS_ATTR_SET(zp, ZFS_NODUMP, xoap->xoa_nodump,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_NODUMP);
}
if (XVA_ISSET_REQ(xvap, XAT_OPAQUE)) {
ZFS_ATTR_SET(zp, ZFS_OPAQUE, xoap->xoa_opaque,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_OPAQUE);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_QUARANTINED)) {
ZFS_ATTR_SET(zp, ZFS_AV_QUARANTINED,
xoap->xoa_av_quarantined, zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_AV_QUARANTINED);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_MODIFIED)) {
ZFS_ATTR_SET(zp, ZFS_AV_MODIFIED, xoap->xoa_av_modified,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_AV_MODIFIED);
}
if (XVA_ISSET_REQ(xvap, XAT_AV_SCANSTAMP)) {
zfs_sa_set_scanstamp(zp, xvap, tx);
XVA_SET_RTN(xvap, XAT_AV_SCANSTAMP);
}
if (XVA_ISSET_REQ(xvap, XAT_REPARSE)) {
ZFS_ATTR_SET(zp, ZFS_REPARSE, xoap->xoa_reparse,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_REPARSE);
}
if (XVA_ISSET_REQ(xvap, XAT_OFFLINE)) {
ZFS_ATTR_SET(zp, ZFS_OFFLINE, xoap->xoa_offline,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_OFFLINE);
}
if (XVA_ISSET_REQ(xvap, XAT_SPARSE)) {
ZFS_ATTR_SET(zp, ZFS_SPARSE, xoap->xoa_sparse,
zp->z_pflags, tx);
XVA_SET_RTN(xvap, XAT_SPARSE);
}
}
/*
* Create a new DMU object to hold a zfs znode.
*
* IN: dzp - parent directory for new znode
* vap - file attributes for new znode
* tx - dmu transaction id for zap operations
* cr - credentials of caller
* flag - flags:
* IS_ROOT_NODE - new object will be root
* IS_XATTR - new object is an attribute
* bonuslen - length of bonus buffer
* setaclp - File/Dir initial ACL
* fuidp - Tracks fuid allocation.
*
* OUT: zpp - allocated znode
*
*/
void
zfs_mknode(znode_t *dzp, vattr_t *vap, dmu_tx_t *tx, cred_t *cr,
uint_t flag, znode_t **zpp, zfs_acl_ids_t *acl_ids)
{
uint64_t crtime[2], atime[2], mtime[2], ctime[2];
uint64_t mode, size, links, parent, pflags;
uint64_t dzp_pflags = 0;
uint64_t rdev = 0;
zfs_sb_t *zsb = ZTOZSB(dzp);
dmu_buf_t *db;
timestruc_t now;
uint64_t gen, obj;
int bonuslen;
sa_handle_t *sa_hdl;
dmu_object_type_t obj_type;
sa_bulk_attr_t *sa_attrs;
int cnt = 0;
zfs_acl_locator_cb_t locate = { 0 };
if (zsb->z_replay) {
obj = vap->va_nodeid;
now = vap->va_ctime; /* see zfs_replay_create() */
gen = vap->va_nblocks; /* ditto */
} else {
obj = 0;
gethrestime(&now);
gen = dmu_tx_get_txg(tx);
}
obj_type = zsb->z_use_sa ? DMU_OT_SA : DMU_OT_ZNODE;
bonuslen = (obj_type == DMU_OT_SA) ?
DN_MAX_BONUSLEN : ZFS_OLD_ZNODE_PHYS_SIZE;
/*
* Create a new DMU object.
*/
/*
* There's currently no mechanism for pre-reading the blocks that will
* be needed to allocate a new object, so we accept the small chance
* that there will be an i/o error and we will fail one of the
* assertions below.
*/
if (S_ISDIR(vap->va_mode)) {
if (zsb->z_replay) {
VERIFY0(zap_create_claim_norm(zsb->z_os, obj,
zsb->z_norm, DMU_OT_DIRECTORY_CONTENTS,
obj_type, bonuslen, tx));
} else {
obj = zap_create_norm(zsb->z_os,
zsb->z_norm, DMU_OT_DIRECTORY_CONTENTS,
obj_type, bonuslen, tx);
}
} else {
if (zsb->z_replay) {
VERIFY0(dmu_object_claim(zsb->z_os, obj,
DMU_OT_PLAIN_FILE_CONTENTS, 0,
obj_type, bonuslen, tx));
} else {
obj = dmu_object_alloc(zsb->z_os,
DMU_OT_PLAIN_FILE_CONTENTS, 0,
obj_type, bonuslen, tx);
}
}
ZFS_OBJ_HOLD_ENTER(zsb, obj);
VERIFY(0 == sa_buf_hold(zsb->z_os, obj, NULL, &db));
/*
* If this is the root, fix up the half-initialized parent pointer
* to reference the just-allocated physical data area.
*/
if (flag & IS_ROOT_NODE) {
dzp->z_id = obj;
} else {
dzp_pflags = dzp->z_pflags;
}
/*
* If parent is an xattr, so am I.
*/
if (dzp_pflags & ZFS_XATTR) {
flag |= IS_XATTR;
}
if (zsb->z_use_fuids)
pflags = ZFS_ARCHIVE | ZFS_AV_MODIFIED;
else
pflags = 0;
if (S_ISDIR(vap->va_mode)) {
size = 2; /* contents ("." and "..") */
links = (flag & (IS_ROOT_NODE | IS_XATTR)) ? 2 : 1;
} else {
size = links = 0;
}
if (S_ISBLK(vap->va_mode) || S_ISCHR(vap->va_mode))
rdev = vap->va_rdev;
parent = dzp->z_id;
mode = acl_ids->z_mode;
if (flag & IS_XATTR)
pflags |= ZFS_XATTR;
/*
* No execs denied will be deterimed when zfs_mode_compute() is called.
*/
pflags |= acl_ids->z_aclp->z_hints &
(ZFS_ACL_TRIVIAL|ZFS_INHERIT_ACE|ZFS_ACL_AUTO_INHERIT|
ZFS_ACL_DEFAULTED|ZFS_ACL_PROTECTED);
ZFS_TIME_ENCODE(&now, crtime);
ZFS_TIME_ENCODE(&now, ctime);
if (vap->va_mask & ATTR_ATIME) {
ZFS_TIME_ENCODE(&vap->va_atime, atime);
} else {
ZFS_TIME_ENCODE(&now, atime);
}
if (vap->va_mask & ATTR_MTIME) {
ZFS_TIME_ENCODE(&vap->va_mtime, mtime);
} else {
ZFS_TIME_ENCODE(&now, mtime);
}
/* Now add in all of the "SA" attributes */
VERIFY(0 == sa_handle_get_from_db(zsb->z_os, db, NULL, SA_HDL_SHARED,
&sa_hdl));
/*
* Setup the array of attributes to be replaced/set on the new file
*
* order for DMU_OT_ZNODE is critical since it needs to be constructed
* in the old znode_phys_t format. Don't change this ordering
*/
sa_attrs = kmem_alloc(sizeof (sa_bulk_attr_t) * ZPL_END, KM_SLEEP);
if (obj_type == DMU_OT_ZNODE) {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ATIME(zsb),
NULL, &atime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MTIME(zsb),
NULL, &mtime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CTIME(zsb),
NULL, &ctime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CRTIME(zsb),
NULL, &crtime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GEN(zsb),
NULL, &gen, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MODE(zsb),
NULL, &mode, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_SIZE(zsb),
NULL, &size, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PARENT(zsb),
NULL, &parent, 8);
} else {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MODE(zsb),
NULL, &mode, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_SIZE(zsb),
NULL, &size, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GEN(zsb),
NULL, &gen, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_UID(zsb),
NULL, &acl_ids->z_fuid, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GID(zsb),
NULL, &acl_ids->z_fgid, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PARENT(zsb),
NULL, &parent, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_FLAGS(zsb),
NULL, &pflags, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ATIME(zsb),
NULL, &atime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_MTIME(zsb),
NULL, &mtime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CTIME(zsb),
NULL, &ctime, 16);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_CRTIME(zsb),
NULL, &crtime, 16);
}
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_LINKS(zsb), NULL, &links, 8);
if (obj_type == DMU_OT_ZNODE) {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_XATTR(zsb), NULL,
&empty_xattr, 8);
}
if (obj_type == DMU_OT_ZNODE ||
(S_ISBLK(vap->va_mode) || S_ISCHR(vap->va_mode))) {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_RDEV(zsb),
NULL, &rdev, 8);
}
if (obj_type == DMU_OT_ZNODE) {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_FLAGS(zsb),
NULL, &pflags, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_UID(zsb), NULL,
&acl_ids->z_fuid, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_GID(zsb), NULL,
&acl_ids->z_fgid, 8);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_PAD(zsb), NULL, pad,
sizeof (uint64_t) * 4);
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_ZNODE_ACL(zsb), NULL,
&acl_phys, sizeof (zfs_acl_phys_t));
} else if (acl_ids->z_aclp->z_version >= ZFS_ACL_VERSION_FUID) {
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_DACL_COUNT(zsb), NULL,
&acl_ids->z_aclp->z_acl_count, 8);
locate.cb_aclp = acl_ids->z_aclp;
SA_ADD_BULK_ATTR(sa_attrs, cnt, SA_ZPL_DACL_ACES(zsb),
zfs_acl_data_locator, &locate,
acl_ids->z_aclp->z_acl_bytes);
mode = zfs_mode_compute(mode, acl_ids->z_aclp, &pflags,
acl_ids->z_fuid, acl_ids->z_fgid);
}
VERIFY(sa_replace_all_by_template(sa_hdl, sa_attrs, cnt, tx) == 0);
if (!(flag & IS_ROOT_NODE)) {
*zpp = zfs_znode_alloc(zsb, db, 0, obj_type, obj, sa_hdl,
ZTOI(dzp));
VERIFY(*zpp != NULL);
VERIFY(dzp != NULL);
} else {
/*
* If we are creating the root node, the "parent" we
* passed in is the znode for the root.
*/
*zpp = dzp;
(*zpp)->z_sa_hdl = sa_hdl;
}
(*zpp)->z_pflags = pflags;
(*zpp)->z_mode = mode;
if (obj_type == DMU_OT_ZNODE ||
acl_ids->z_aclp->z_version < ZFS_ACL_VERSION_FUID) {
VERIFY0(zfs_aclset_common(*zpp, acl_ids->z_aclp, cr, tx));
}
kmem_free(sa_attrs, sizeof (sa_bulk_attr_t) * ZPL_END);
ZFS_OBJ_HOLD_EXIT(zsb, obj);
}
/*
* Link zp into dl. Can only fail if zp has been unlinked.
*/
int
zfs_link_create(zfs_dirlock_t *dl, znode_t *zp, dmu_tx_t *tx, int flag)
{
znode_t *dzp = dl->dl_dzp;
zfs_sb_t *zsb = ZTOZSB(zp);
uint64_t value;
int zp_is_dir = S_ISDIR(ZTOI(zp)->i_mode);
sa_bulk_attr_t bulk[5];
uint64_t mtime[2], ctime[2];
int count = 0;
int error;
mutex_enter(&zp->z_lock);
if (!(flag & ZRENAMING)) {
if (zp->z_unlinked) { /* no new links to unlinked zp */
ASSERT(!(flag & (ZNEW | ZEXISTS)));
mutex_exit(&zp->z_lock);
return (ENOENT);
}
zp->z_links++;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zsb), NULL,
&zp->z_links, sizeof (zp->z_links));
}
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_PARENT(zsb), NULL,
&dzp->z_id, sizeof (dzp->z_id));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zsb), NULL,
&zp->z_pflags, sizeof (zp->z_pflags));
if (!(flag & ZNEW)) {
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zsb), NULL,
ctime, sizeof (ctime));
zfs_tstamp_update_setup(zp, STATE_CHANGED, mtime,
ctime, B_TRUE);
}
error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
mutex_exit(&zp->z_lock);
mutex_enter(&dzp->z_lock);
dzp->z_size++;
dzp->z_links += zp_is_dir;
count = 0;
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zsb), NULL,
&dzp->z_size, sizeof (dzp->z_size));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zsb), NULL,
&dzp->z_links, sizeof (dzp->z_links));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zsb), NULL,
mtime, sizeof (mtime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zsb), NULL,
ctime, sizeof (ctime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zsb), NULL,
&dzp->z_pflags, sizeof (dzp->z_pflags));
zfs_tstamp_update_setup(dzp, CONTENT_MODIFIED, mtime, ctime, B_TRUE);
error = sa_bulk_update(dzp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
mutex_exit(&dzp->z_lock);
value = zfs_dirent(zp, zp->z_mode);
error = zap_add(ZTOZSB(zp)->z_os, dzp->z_id, dl->dl_name,
8, 1, &value, tx);
ASSERT(error == 0);
return (0);
}
int
zfs_rezget(znode_t *zp)
{
zfs_sb_t *zsb = ZTOZSB(zp);
dmu_object_info_t doi;
dmu_buf_t *db;
uint64_t obj_num = zp->z_id;
uint64_t mode;
sa_bulk_attr_t bulk[8];
int err;
int count = 0;
uint64_t gen;
ZFS_OBJ_HOLD_ENTER(zsb, obj_num);
mutex_enter(&zp->z_acl_lock);
if (zp->z_acl_cached) {
zfs_acl_free(zp->z_acl_cached);
zp->z_acl_cached = NULL;
}
mutex_exit(&zp->z_acl_lock);
rw_enter(&zp->z_xattr_lock, RW_WRITER);
if (zp->z_xattr_cached) {
nvlist_free(zp->z_xattr_cached);
zp->z_xattr_cached = NULL;
}
if (zp->z_xattr_parent) {
iput(ZTOI(zp->z_xattr_parent));
zp->z_xattr_parent = NULL;
}
rw_exit(&zp->z_xattr_lock);
ASSERT(zp->z_sa_hdl == NULL);
err = sa_buf_hold(zsb->z_os, obj_num, NULL, &db);
if (err) {
ZFS_OBJ_HOLD_EXIT(zsb, obj_num);
return (err);
}
dmu_object_info_from_db(db, &doi);
if (doi.doi_bonus_type != DMU_OT_SA &&
(doi.doi_bonus_type != DMU_OT_ZNODE ||
(doi.doi_bonus_type == DMU_OT_ZNODE &&
doi.doi_bonus_size < sizeof (znode_phys_t)))) {
sa_buf_rele(db, NULL);
ZFS_OBJ_HOLD_EXIT(zsb, obj_num);
return (SET_ERROR(EINVAL));
}
zfs_znode_sa_init(zsb, zp, db, doi.doi_bonus_type, NULL);
/* reload cached values */
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GEN(zsb), NULL,
&gen, sizeof (gen));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_SIZE(zsb), NULL,
&zp->z_size, sizeof (zp->z_size));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zsb), NULL,
&zp->z_links, sizeof (zp->z_links));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zsb), NULL,
&zp->z_pflags, sizeof (zp->z_pflags));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_ATIME(zsb), NULL,
&zp->z_atime, sizeof (zp->z_atime));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_UID(zsb), NULL,
&zp->z_uid, sizeof (zp->z_uid));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_GID(zsb), NULL,
&zp->z_gid, sizeof (zp->z_gid));
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MODE(zsb), NULL,
&mode, sizeof (mode));
if (sa_bulk_lookup(zp->z_sa_hdl, bulk, count)) {
zfs_znode_dmu_fini(zp);
ZFS_OBJ_HOLD_EXIT(zsb, obj_num);
return (SET_ERROR(EIO));
}
zp->z_mode = mode;
if (gen != zp->z_gen) {
zfs_znode_dmu_fini(zp);
ZFS_OBJ_HOLD_EXIT(zsb, obj_num);
return (SET_ERROR(EIO));
}
zp->z_unlinked = (zp->z_links == 0);
zp->z_blksz = doi.doi_data_block_size;
zfs_inode_update(zp);
ZFS_OBJ_HOLD_EXIT(zsb, obj_num);
return (0);
}
/*
* Free space in a file
*
* IN: zp - znode of file to free data in.
* off - start of range
* len - end of range (0 => EOF)
* flag - current file open mode flags.
* log - TRUE if this action should be logged
*
* RETURN: 0 on success, error code on failure
*/
int
zfs_freesp(znode_t *zp, uint64_t off, uint64_t len, int flag, boolean_t log)
{
struct inode *ip = ZTOI(zp);
dmu_tx_t *tx;
zfs_sb_t *zsb = ZTOZSB(zp);
zilog_t *zilog = zsb->z_log;
uint64_t mode;
uint64_t mtime[2], ctime[2];
sa_bulk_attr_t bulk[3];
int count = 0;
int error;
if ((error = sa_lookup(zp->z_sa_hdl, SA_ZPL_MODE(zsb), &mode,
sizeof (mode))) != 0)
return (error);
if (off > zp->z_size) {
error = zfs_extend(zp, off+len);
if (error == 0 && log)
goto log;
else
return (error);
}
/*
* Check for any locks in the region to be freed.
*/
if (ip->i_flock && mandatory_lock(ip)) {
uint64_t length = (len ? len : zp->z_size - off);
if (!lock_may_write(ip, off, length))
return (SET_ERROR(EAGAIN));
}
if (len == 0) {
error = zfs_trunc(zp, off);
} else {
if ((error = zfs_free_range(zp, off, len)) == 0 &&
off + len > zp->z_size)
error = zfs_extend(zp, off+len);
}
if (error || !log)
return (error);
log:
tx = dmu_tx_create(zsb->z_os);
dmu_tx_hold_sa(tx, zp->z_sa_hdl, B_FALSE);
zfs_sa_upgrade_txholds(tx, zp);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
dmu_tx_abort(tx);
return (error);
}
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zsb), NULL, mtime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zsb), NULL, ctime, 16);
SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_FLAGS(zsb),
NULL, &zp->z_pflags, 8);
zfs_tstamp_update_setup(zp, CONTENT_MODIFIED, mtime, ctime, B_TRUE);
error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
ASSERT(error == 0);
zfs_log_truncate(zilog, tx, TX_TRUNCATE, zp, off, len);
dmu_tx_commit(tx);
zfs_inode_update(zp);
return (0);
}
void
zfs_log_write(zilog_t *zilog, dmu_tx_t *tx, int txtype,
znode_t *zp, offset_t off, ssize_t resid, int ioflag,
zil_callback_t callback, void *callback_data)
{
itx_wr_state_t write_state;
boolean_t slogging;
uintptr_t fsync_cnt;
ssize_t immediate_write_sz;
if (zil_replaying(zilog, tx) || zp->z_unlinked) {
if (callback != NULL)
callback(callback_data);
return;
}
immediate_write_sz = (zilog->zl_logbias == ZFS_LOGBIAS_THROUGHPUT)
? 0 : (ssize_t)zfs_immediate_write_sz;
slogging = spa_has_slogs(zilog->zl_spa) &&
(zilog->zl_logbias == ZFS_LOGBIAS_LATENCY);
if (resid > immediate_write_sz && !slogging && resid <= zp->z_blksz)
write_state = WR_INDIRECT;
else if (ioflag & (FSYNC | FDSYNC))
write_state = WR_COPIED;
else
write_state = WR_NEED_COPY;
if ((fsync_cnt = (uintptr_t)tsd_get(zfs_fsyncer_key)) != 0) {
(void) tsd_set(zfs_fsyncer_key, (void *)(fsync_cnt - 1));
}
while (resid) {
itx_t *itx;
lr_write_t *lr;
ssize_t len;
/*
* If the write would overflow the largest block then split it.
*/
if (write_state != WR_INDIRECT && resid > ZIL_MAX_LOG_DATA)
len = SPA_MAXBLOCKSIZE >> 1;
else
len = resid;
itx = zil_itx_create(txtype, sizeof (*lr) +
(write_state == WR_COPIED ? len : 0));
lr = (lr_write_t *)&itx->itx_lr;
if (write_state == WR_COPIED && dmu_read(ZTOZSB(zp)->z_os,
zp->z_id, off, len, lr + 1, DMU_READ_NO_PREFETCH) != 0) {
zil_itx_destroy(itx);
itx = zil_itx_create(txtype, sizeof (*lr));
lr = (lr_write_t *)&itx->itx_lr;
write_state = WR_NEED_COPY;
}
itx->itx_wr_state = write_state;
if (write_state == WR_NEED_COPY)
itx->itx_sod += len;
lr->lr_foid = zp->z_id;
lr->lr_offset = off;
lr->lr_length = len;
lr->lr_blkoff = 0;
BP_ZERO(&lr->lr_blkptr);
itx->itx_private = ZTOZSB(zp);
if (!(ioflag & (FSYNC | FDSYNC)) && (zp->z_sync_cnt == 0) &&
(fsync_cnt == 0))
itx->itx_sync = B_FALSE;
itx->itx_callback = callback;
itx->itx_callback_data = callback_data;
zil_itx_assign(zilog, itx, tx);
off += len;
resid -= len;
}
void
zfs_fuid_map_ids(znode_t *zp, cred_t *cr, uid_t *uidp, uid_t *gidp)
{
*uidp = zfs_fuid_map_id(ZTOZSB(zp), zp->z_uid, cr, ZFS_OWNER);
*gidp = zfs_fuid_map_id(ZTOZSB(zp), zp->z_gid, cr, ZFS_GROUP);
}
