/*
* Handles TX_SYMLINK transactions.
*/
void
zfs_log_symlink(zilog_t *zilog, dmu_tx_t *tx, uint64_t txtype,
znode_t *dzp, znode_t *zp, char *name, char *link)
{
itx_t *itx;
lr_create_t *lr;
size_t namesize = strlen(name) + 1;
size_t linksize = strlen(link) + 1;
if (zil_replaying(zilog, tx))
return;
itx = zil_itx_create(txtype, sizeof (*lr) + namesize + linksize);
lr = (lr_create_t *)&itx->itx_lr;
lr->lr_doid = dzp->z_id;
lr->lr_foid = zp->z_id;
lr->lr_uid = zp->z_uid;
lr->lr_gid = zp->z_gid;
lr->lr_mode = zp->z_mode;
(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);
bcopy(name, (char *)(lr + 1), namesize);
bcopy(link, (char *)(lr + 1) + namesize, linksize);
zil_itx_assign(zilog, itx, tx);
}
void
zfs_rmnode(znode_t *zp)
{
zfsvfs_t *zfsvfs = zp->z_zfsvfs;
objset_t *os = zfsvfs->z_os;
znode_t *xzp = NULL;
dmu_tx_t *tx;
uint64_t acl_obj;
uint64_t xattr_obj;
int error;
ASSERT(zp->z_links == 0);
ASSERT_VOP_ELOCKED(ZTOV(zp), __func__);
/*
* If this is an attribute directory, purge its contents.
*/
if (ZTOV(zp) != NULL && ZTOV(zp)->v_type == VDIR &&
(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);
zfs_znode_free(zp);
return;
}
} else {
/*
* Free up all the data in the file. We don't do this for
* XATTR 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.
*/
error = dmu_free_long_range(os, zp->z_id, 0, DMU_OBJECT_END);
if (error) {
/*
* Not enough space. Leave the file in the unlinked
* set.
*/
zfs_znode_dmu_fini(zp);
zfs_znode_free(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);
ASSERT3S(error, ==, 0);
vn_lock(ZTOV(xzp), LK_EXCLUSIVE | LK_RETRY);
}
/*
* Copy an extended attribute into the buffer provided, or compute the
* required buffer size.
*
* If buf is NULL, it computes the required buffer size.
*
* Returns 0 on success or a negative error number on failure.
* On success, the number of bytes used / required is stored in 'size'.
*
* No locking is done here.
*/
int __osd_xattr_load(udmu_objset_t *uos, uint64_t dnode, nvlist_t **sa_xattr)
{
sa_handle_t *sa_hdl;
char *buf;
int rc, size;
if (unlikely(dnode == ZFS_NO_OBJECT))
return -ENOENT;
rc = -sa_handle_get(uos->os, dnode, NULL, SA_HDL_PRIVATE, &sa_hdl);
if (rc)
return rc;
rc = -sa_size(sa_hdl, SA_ZPL_DXATTR(uos), &size);
if (rc) {
if (rc == -ENOENT)
rc = -nvlist_alloc(sa_xattr, NV_UNIQUE_NAME, KM_SLEEP);
goto out_sa;
}
buf = sa_spill_alloc(KM_SLEEP);
if (buf == NULL) {
rc = -ENOMEM;
goto out_sa;
}
rc = -sa_lookup(sa_hdl, SA_ZPL_DXATTR(uos), buf, size);
if (rc == 0)
rc = -nvlist_unpack(buf, size, sa_xattr, KM_SLEEP);
sa_spill_free(buf);
out_sa:
sa_handle_destroy(sa_hdl);
return rc;
}
static void
ui_delete(char *cmd)
{
char cookies_str[ISAKMP_HDR_COOKIES_LEN * 2 + 1];
char message_id_str[ISAKMP_HDR_MESSAGE_ID_LEN * 2 + 1];
u_int8_t cookies[ISAKMP_HDR_COOKIES_LEN];
u_int8_t message_id_buf[ISAKMP_HDR_MESSAGE_ID_LEN];
u_int8_t *message_id = message_id_buf;
struct sa *sa;
if (sscanf(cmd, "d %32s %8s", cookies_str, message_id_str) != 2) {
log_print("ui_delete: command \"%s\" malformed", cmd);
return;
}
if (strcmp(message_id_str, "-") == 0)
message_id = 0;
if (hex2raw(cookies_str, cookies, ISAKMP_HDR_COOKIES_LEN) == -1 ||
(message_id && hex2raw(message_id_str, message_id_buf,
ISAKMP_HDR_MESSAGE_ID_LEN) == -1)) {
log_print("ui_delete: command \"%s\" has bad arguments", cmd);
return;
}
sa = sa_lookup(cookies, message_id);
if (!sa) {
log_print("ui_delete: command \"%s\" found no SA", cmd);
return;
}
LOG_DBG((LOG_UI, 20,
"ui_delete: deleting SA for cookie \"%s\" msgid \"%s\"",
cookies_str, message_id_str));
sa_delete(sa, 1);
}
int
zfs_sa_get_xattr(znode_t *zp)
{
zfs_sb_t *zsb = ZTOZSB(zp);
char *obj;
int size;
int error;
ASSERT(RW_LOCK_HELD(&zp->z_xattr_lock));
ASSERT(!zp->z_xattr_cached);
ASSERT(zp->z_is_sa);
error = sa_size(zp->z_sa_hdl, SA_ZPL_DXATTR(zsb), &size);
if (error) {
if (error == ENOENT)
return nvlist_alloc(&zp->z_xattr_cached,
NV_UNIQUE_NAME, KM_SLEEP);
else
return (error);
}
obj = zio_buf_alloc(size);
error = sa_lookup(zp->z_sa_hdl, SA_ZPL_DXATTR(zsb), obj, size);
if (error == 0)
error = nvlist_unpack(obj, size, &zp->z_xattr_cached, KM_SLEEP);
zio_buf_free(obj, size);
return (error);
}
int __osd_xattr_load(struct osd_device *osd, uint64_t dnode, nvlist_t **sa)
{
sa_handle_t *sa_hdl;
char *buf;
int rc, size;
if (unlikely(dnode == ZFS_NO_OBJECT))
return -ENOENT;
rc = -sa_handle_get(osd->od_os, dnode, NULL, SA_HDL_PRIVATE, &sa_hdl);
if (rc)
return rc;
rc = -sa_size(sa_hdl, SA_ZPL_DXATTR(osd), &size);
if (rc) {
if (rc == -ENOENT)
rc = -nvlist_alloc(sa, NV_UNIQUE_NAME, KM_SLEEP);
goto out_sa;
}
buf = osd_zio_buf_alloc(size);
if (buf == NULL) {
rc = -ENOMEM;
goto out_sa;
}
rc = -sa_lookup(sa_hdl, SA_ZPL_DXATTR(osd), buf, size);
if (rc == 0)
rc = -nvlist_unpack(buf, size, sa, KM_SLEEP);
osd_zio_buf_free(buf, size);
out_sa:
sa_handle_destroy(sa_hdl);
return rc;
}
void
zfs_sa_get_scanstamp(znode_t *zp, xvattr_t *xvap)
{
zfsvfs_t *zfsvfs = zp->z_zfsvfs;
xoptattr_t *xoap;
ASSERT(MUTEX_HELD(&zp->z_lock));
VERIFY((xoap = xva_getxoptattr(xvap)) != NULL);
if (zp->z_is_sa) {
if (sa_lookup(zp->z_sa_hdl, SA_ZPL_SCANSTAMP(zfsvfs),
&xoap->xoa_av_scanstamp,
sizeof (xoap->xoa_av_scanstamp)) != 0)
return;
} else {
dmu_object_info_t doi;
dmu_buf_t *db = sa_get_db(zp->z_sa_hdl);
int len;
if (!(zp->z_pflags & ZFS_BONUS_SCANSTAMP))
return;
sa_object_info(zp->z_sa_hdl, &doi);
len = sizeof (xoap->xoa_av_scanstamp) +
ZFS_OLD_ZNODE_PHYS_SIZE;
if (len <= doi.doi_bonus_size) {
(void) memcpy(xoap->xoa_av_scanstamp,
(caddr_t)db->db_data + ZFS_OLD_ZNODE_PHYS_SIZE,
sizeof (xoap->xoa_av_scanstamp));
}
}
XVA_SET_RTN(xvap, XAT_AV_SCANSTAMP);
}
/*
* Look up an entry in a directory.
*
* NOTE: '.' and '..' are handled as special cases because
* no directory entries are actually stored for them. If this is
* the root of a filesystem, then '.zfs' is also treated as a
* special pseudo-directory.
*/
int
zfs_dirlook(znode_t *dzp, char *name, struct inode **ipp, int flags,
int *deflg, pathname_t *rpnp)
{
zfs_dirlock_t *dl;
znode_t *zp;
int error = 0;
uint64_t parent;
if (name[0] == 0 || (name[0] == '.' && name[1] == 0)) {
*ipp = ZTOI(dzp);
igrab(*ipp);
} else if (name[0] == '.' && name[1] == '.' && name[2] == 0) {
zfs_sb_t *zsb = ZTOZSB(dzp);
/*
* If we are a snapshot mounted under .zfs, return
* the vp for the snapshot directory.
*/
if ((error = sa_lookup(dzp->z_sa_hdl,
SA_ZPL_PARENT(zsb), &parent, sizeof (parent))) != 0)
return (error);
#ifdef HAVE_SNAPSHOT
if (parent == dzp->z_id && zsb->z_parent != zsb) {
error = zfsctl_root_lookup(zsb->z_parent->z_ctldir,
"snapshot", ipp, NULL, 0, NULL, kcred,
NULL, NULL, NULL);
return (error);
}
#endif /* HAVE_SNAPSHOT */
rw_enter(&dzp->z_parent_lock, RW_READER);
error = zfs_zget(zsb, parent, &zp);
if (error == 0)
*ipp = ZTOI(zp);
rw_exit(&dzp->z_parent_lock);
#ifdef HAVE_SNAPSHOT
} else if (zfs_has_ctldir(dzp) && strcmp(name, ZFS_CTLDIR_NAME) == 0) {
*ipp = zfsctl_root(dzp);
#endif /* HAVE_SNAPSHOT */
} else {
int zf;
zf = ZEXISTS | ZSHARED;
if (flags & FIGNORECASE)
zf |= ZCILOOK;
error = zfs_dirent_lock(&dl, dzp, name, &zp, zf, deflg, rpnp);
if (error == 0) {
*ipp = ZTOI(zp);
zfs_dirent_unlock(dl);
dzp->z_zn_prefetch = B_TRUE; /* enable prefetching */
}
rpnp = NULL;
}
if ((flags & FIGNORECASE) && rpnp && !error)
(void) strlcpy(rpnp->pn_buf, name, rpnp->pn_bufsize);
return (error);
}
void dirattrpack(attrinfo_t *aip, znode_t *zp)
{
attrgroup_t dirattr = aip->ai_attrlist->dirattr;
void *attrbufptr = *aip->ai_attrbufpp;
if (ATTR_DIR_LINKCOUNT & dirattr) {
*((u_int32_t *)attrbufptr) = 1; /* no dir hard links */
attrbufptr = ((u_int32_t *)attrbufptr) + 1;
}
if (ATTR_DIR_ENTRYCOUNT & dirattr) {
uint64_t val;
VERIFY(sa_lookup(zp->z_sa_hdl, SA_ZPL_SIZE(zp->z_zfsvfs),
&val, sizeof(val)) == 0);
*((u_int32_t *)attrbufptr) = (uint32_t)val;
attrbufptr = ((u_int32_t *)attrbufptr) + 1;
}
if (ATTR_DIR_MOUNTSTATUS & dirattr && zp) {
vnode_t *vp = ZTOV(zp);
if (vp != NULL && vnode_mountedhere(vp) != NULL)
*((u_int32_t *)attrbufptr) = DIR_MNTSTATUS_MNTPOINT;
else
*((u_int32_t *)attrbufptr) = 0;
attrbufptr = ((u_int32_t *)attrbufptr) + 1;
}
*aip->ai_attrbufpp = attrbufptr;
}
uint32_t
zfs_getbsdflags(znode_t *zp)
{
uint32_t bsdflags = 0;
uint64_t zflags;
VERIFY(sa_lookup(zp->z_sa_hdl, SA_ZPL_FLAGS(zp->z_zfsvfs),
&zflags, sizeof (zflags)) == 0);
if (zflags & ZFS_NODUMP)
bsdflags |= UF_NODUMP;
if (zflags & ZFS_IMMUTABLE)
bsdflags |= UF_IMMUTABLE;
if (zflags & ZFS_APPENDONLY)
bsdflags |= UF_APPEND;
if (zflags & ZFS_OPAQUE)
bsdflags |= UF_OPAQUE;
if (zflags & ZFS_HIDDEN)
bsdflags |= UF_HIDDEN;
/*
* Due to every file getting archive set automatically, and OSX
* don't let you move/copy it as a user, we disable archive connection
* for now
if (zflags & ZFS_ARCHIVE)
bsdflags |= SF_ARCHIVED;
*/
dprintf("getbsd changing zfs %08lx to osx %08lx\n",
zflags, bsdflags);
return (bsdflags);
}
/*
* Create the '.zfs' directory. This directory is cached as part of the VFS
* structure. This results in a hold on the vfs_t. The code in zfs_umount()
* therefore checks against a vfs_count of 2 instead of 1. This reference
* is removed when the ctldir is destroyed in the unmount.
*/
void
zfsctl_create(zfsvfs_t *zfsvfs)
{
vnode_t *vp, *rvp;
zfsctl_node_t *zcp;
uint64_t crtime[2];
ASSERT(zfsvfs->z_ctldir == NULL);
vp = gfs_root_create(sizeof (zfsctl_node_t), zfsvfs->z_vfs,
&zfsctl_ops_root, ZFSCTL_INO_ROOT, zfsctl_root_entries,
zfsctl_root_inode_cb, MAXNAMELEN, NULL, NULL);
zcp = vp->v_data;
zcp->zc_id = ZFSCTL_INO_ROOT;
VERIFY(VFS_ROOT(zfsvfs->z_vfs, LK_EXCLUSIVE, &rvp) == 0);
VERIFY(0 == sa_lookup(VTOZ(rvp)->z_sa_hdl, SA_ZPL_CRTIME(zfsvfs),
&crtime, sizeof (crtime)));
ZFS_TIME_DECODE(&zcp->zc_cmtime, crtime);
VN_URELE(rvp);
/*
* We're only faking the fact that we have a root of a filesystem for
* the sake of the GFS interfaces. Undo the flag manipulation it did
* for us.
*/
vp->v_vflag &= ~VV_ROOT;
zfsvfs->z_ctldir = vp;
VOP_UNLOCK(vp, 0);
}
int __osd_xattr_get_large(const struct lu_env *env, struct osd_device *osd,
uint64_t xattr, struct lu_buf *buf,
const char *name, int *sizep)
{
dmu_buf_t *xa_data_db;
sa_handle_t *sa_hdl = NULL;
uint64_t xa_data_obj, size;
int rc;
/* are there any extended attributes? */
if (xattr == ZFS_NO_OBJECT)
return -ENOENT;
/* Lookup the object number containing the xattr data */
rc = -zap_lookup(osd->od_os, xattr, name, sizeof(uint64_t), 1,
&xa_data_obj);
if (rc)
return rc;
rc = __osd_obj2dbuf(env, osd->od_os, xa_data_obj, &xa_data_db);
if (rc)
return rc;
rc = -sa_handle_get(osd->od_os, xa_data_obj, NULL, SA_HDL_PRIVATE,
&sa_hdl);
if (rc)
goto out_rele;
/* Get the xattr value length / object size */
rc = -sa_lookup(sa_hdl, SA_ZPL_SIZE(osd), &size, 8);
if (rc)
goto out;
if (size > INT_MAX) {
rc = -EOVERFLOW;
goto out;
}
*sizep = (int)size;
if (buf == NULL || buf->lb_buf == NULL) {
/* We only need to return the required size */
goto out;
}
if (*sizep > buf->lb_len) {
rc = -ERANGE; /* match ldiskfs error */
goto out;
}
rc = -dmu_read(osd->od_os, xa_data_db->db_object, 0,
size, buf->lb_buf, DMU_READ_PREFETCH);
out:
sa_handle_destroy(sa_hdl);
out_rele:
dmu_buf_rele(xa_data_db, FTAG);
return rc;
}
void getfinderinfo(znode_t *zp, cred_t *cr, finderinfo_t *fip)
{
vnode_t *xdvp = NULLVP;
vnode_t *xvp = NULLVP;
struct uio *auio = NULL;
struct componentname cn;
int error;
uint64_t xattr = 0;
if (sa_lookup(zp->z_sa_hdl, SA_ZPL_XATTR(zp->z_zfsvfs),
&xattr, sizeof(xattr)) ||
(xattr == 0)) {
goto nodata;
}
auio = uio_create(1, 0, UIO_SYSSPACE, UIO_READ);
if (auio == NULL) {
goto nodata;
}
uio_addiov(auio, CAST_USER_ADDR_T(fip), sizeof (finderinfo_t));
/*
* Grab the hidden attribute directory vnode.
*
* XXX - switch to embedded Finder Info when it becomes available
*/
if ((error = zfs_get_xattrdir(zp, &xdvp, cr, 0))) {
goto out;
}
bzero(&cn, sizeof (cn));
cn.cn_nameiop = LOOKUP;
cn.cn_flags = ISLASTCN;
cn.cn_nameptr = XATTR_FINDERINFO_NAME;
cn.cn_namelen = strlen(cn.cn_nameptr);
if ((error = zfs_dirlook(VTOZ(xdvp), cn.cn_nameptr, &xvp, 0, NULL, &cn))) {
goto out;
}
error = dmu_read_uio(zp->z_zfsvfs->z_os, VTOZ(xvp)->z_id, auio,
sizeof (finderinfo_t));
out:
if (auio)
uio_free(auio);
if (xvp)
vnode_put(xvp);
if (xdvp)
vnode_put(xdvp);
if (error == 0)
return;
nodata:
bzero(fip, sizeof (finderinfo_t));
}
zfs_dirlook(znode_t *dzp, char *name, vnode_t **vpp)
#endif
{
zfs_dirlock_t *dl;
znode_t *zp;
int error = 0;
uint64_t parent;
#ifdef __APPLE__
char *name = cnp->cn_nameptr;
#endif
if (name[0] == 0 || (name[0] == '.' && name[1] == 0)) {
*vpp = ZTOV(dzp);
VN_HOLD(*vpp);
} else if (name[0] == '.' && name[1] == '.' && name[2] == 0) {
zfsvfs_t *zfsvfs = dzp->z_zfsvfs;
/*
* If we are a snapshot mounted under .zfs, return
* the vp for the snapshot directory.
*/
if ((error = sa_lookup(dzp->z_sa_hdl,
SA_ZPL_PARENT(zfsvfs), &parent, sizeof (parent))) != 0)
return (error);
if (parent == dzp->z_id && zfsvfs->z_parent != zfsvfs) {
error = zfsctl_root_lookup(zfsvfs->z_parent->z_ctldir,
"snapshot", vpp, NULL, 0, NULL, kcred
/*, NULL, NULL, NULL*/);
return (error);
}
rw_enter(&dzp->z_parent_lock, RW_READER);
error = zfs_zget(zfsvfs, parent, &zp);
if (error == 0)
*vpp = ZTOV(zp);
rw_exit(&dzp->z_parent_lock);
} else if (zfs_has_ctldir(dzp) && strcmp(name, ZFS_CTLDIR_NAME) == 0) {
*vpp = zfsctl_root(dzp);
} else {
#ifdef __APPLE__
error = zfs_dirent_lock(&dl, dzp, cnp, &zp, ZEXISTS | ZSHARED);
#else
error = zfs_dirent_lock(&dl, dzp, name, &zp, ZEXISTS | ZSHARED);
#endif
if (error == 0) {
*vpp = ZTOV(zp);
zfs_dirent_unlock(dl);
dzp->z_zn_prefetch = B_TRUE; /* enable prefetching */
}
}
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);
}
/*
* Compute the same user access value as getattrlist(2)
*/
u_int32_t getuseraccess(znode_t *zp, vfs_context_t ctx)
{
vnode_t *vp;
u_int32_t user_access = 0;
zfs_acl_phys_t acl_phys;
int error;
/* Only take the expensive vnode_authorize path when we have an ACL */
error = sa_lookup(zp->z_sa_hdl, SA_ZPL_ZNODE_ACL(zp->z_zfsvfs),
&acl_phys, sizeof (acl_phys));
if (error || acl_phys.z_acl_count == 0) {
kauth_cred_t cred = vfs_context_ucred(ctx);
uint64_t obj_uid;
uint64_t obj_mode;
/* User id 0 (root) always gets access. */
if (!vfs_context_suser(ctx)) {
return (R_OK | W_OK | X_OK);
}
sa_lookup(zp->z_sa_hdl, SA_ZPL_UID(zp->z_zfsvfs),
&obj_uid, sizeof (obj_uid));
sa_lookup(zp->z_sa_hdl, SA_ZPL_MODE(zp->z_zfsvfs),
&obj_mode, sizeof (obj_mode));
//obj_uid = pzp->zp_uid;
obj_mode = obj_mode & MODEMASK;
if (obj_uid == UNKNOWNUID) {
obj_uid = kauth_cred_getuid(cred);
}
if ((obj_uid == kauth_cred_getuid(cred)) ||
(obj_uid == UNKNOWNUID)) {
return (((u_int32_t)obj_mode & S_IRWXU) >> 6);
}
/* Otherwise, settle for 'others' access. */
return ((u_int32_t)obj_mode & S_IRWXO);
}
void
zfs_setbsdflags(znode_t *zp, uint32_t bsdflags)
{
uint64_t zflags;
VERIFY(sa_lookup(zp->z_sa_hdl, SA_ZPL_FLAGS(zp->z_zfsvfs),
&zflags, sizeof (zflags)) == 0);
if (bsdflags & UF_NODUMP)
zflags |= ZFS_NODUMP;
else
zflags &= ~ZFS_NODUMP;
if (bsdflags & UF_IMMUTABLE)
zflags |= ZFS_IMMUTABLE;
else
zflags &= ~ZFS_IMMUTABLE;
if (bsdflags & UF_APPEND)
zflags |= ZFS_APPENDONLY;
else
zflags &= ~ZFS_APPENDONLY;
if (bsdflags & UF_OPAQUE)
zflags |= ZFS_OPAQUE;
else
zflags &= ~ZFS_OPAQUE;
if (bsdflags & UF_HIDDEN)
zflags |= ZFS_HIDDEN;
else
zflags &= ~ZFS_HIDDEN;
/*
if (bsdflags & SF_ARCHIVED)
zflags |= ZFS_ARCHIVE;
else
zflags &= ~ZFS_ARCHIVE;
*/
zp->z_pflags = zflags;
dprintf("setbsd changing osx %08lx to zfs %08lx\n",
bsdflags, zflags);
/*
(void )sa_update(zp->z_sa_hdl, SA_ZPL_FLAGS(zp->z_zfsvfs),
(void *)&zp->z_pflags, sizeof (uint64_t), tx);
*/
}
static void
zfs_inode_set_ops(zfs_sb_t *zsb, struct inode *ip)
{
uint64_t rdev = 0;
switch (ip->i_mode & S_IFMT) {
case S_IFREG:
ip->i_op = &zpl_inode_operations;
ip->i_fop = &zpl_file_operations;
ip->i_mapping->a_ops = &zpl_address_space_operations;
break;
case S_IFDIR:
ip->i_op = &zpl_dir_inode_operations;
ip->i_fop = &zpl_dir_file_operations;
ITOZ(ip)->z_zn_prefetch = B_TRUE;
break;
case S_IFLNK:
ip->i_op = &zpl_symlink_inode_operations;
break;
/*
* rdev is only stored in a SA only for device files.
*/
case S_IFCHR:
case S_IFBLK:
VERIFY(sa_lookup(ITOZ(ip)->z_sa_hdl, SA_ZPL_RDEV(zsb),
&rdev, sizeof (rdev)) == 0);
/*FALLTHROUGH*/
case S_IFIFO:
case S_IFSOCK:
init_special_inode(ip, ip->i_mode, rdev);
ip->i_op = &zpl_special_inode_operations;
break;
default:
printk("ZFS: Invalid mode: 0x%x\n", ip->i_mode);
VERIFY(0);
}
}
/*
* Copy an extended attribute into the buffer provided, or compute the
* required buffer size.
*
* If buf is NULL, it computes the required buffer size.
*
* Returns 0 on success or a negative error number on failure.
* On success, the number of bytes used / required is stored in 'size'.
*
* No locking is done here.
*/
int __osd_xattr_cache(const struct lu_env *env, struct osd_object *obj)
{
struct osd_device *osd = osd_obj2dev(obj);
udmu_objset_t *uos = &osd->od_objset;
sa_handle_t *sa_hdl;
char *buf;
int size;
int rc;
LASSERT(obj->oo_sa_xattr == NULL);
LASSERT(obj->oo_db != NULL);
rc = -sa_handle_get(uos->os, obj->oo_db->db_object, NULL,
SA_HDL_PRIVATE, &sa_hdl);
if (rc)
return rc;
rc = -sa_size(sa_hdl, SA_ZPL_DXATTR(uos), &size);
if (rc) {
if (rc == -ENOENT)
rc = -nvlist_alloc(&obj->oo_sa_xattr,
NV_UNIQUE_NAME, KM_SLEEP);
goto out_sa;
}
buf = sa_spill_alloc(KM_SLEEP);
if (buf == NULL) {
rc = -ENOMEM;
goto out_sa;
}
rc = -sa_lookup(sa_hdl, SA_ZPL_DXATTR(uos), buf, size);
if (rc == 0)
rc = -nvlist_unpack(buf, size, &obj->oo_sa_xattr, KM_SLEEP);
sa_spill_free(buf);
out_sa:
sa_handle_destroy(sa_hdl);
return rc;
}
static int
zfs_dd_lookup(znode_t *dzp, znode_t **zpp)
{
zfsvfs_t *zfsvfs = dzp->z_zfsvfs;
znode_t *zp;
uint64_t parent;
int error;
ASSERT_VOP_LOCKED(ZTOV(dzp), __func__);
ASSERT(RRM_READ_HELD(&zfsvfs->z_teardown_lock));
if (dzp->z_unlinked)
return (ENOENT);
if ((error = sa_lookup(dzp->z_sa_hdl,
SA_ZPL_PARENT(zfsvfs), &parent, sizeof (parent))) != 0)
return (error);
error = zfs_zget(zfsvfs, parent, &zp);
if (error == 0)
*zpp = zp;
return (error);
}
static int
osd_object_sa_init(struct osd_object *obj, udmu_objset_t *uos)
{
int rc;
LASSERT(obj->oo_sa_hdl == NULL);
LASSERT(obj->oo_db != NULL);
rc = -sa_handle_get(uos->os, obj->oo_db->db_object, obj,
SA_HDL_PRIVATE, &obj->oo_sa_hdl);
if (rc)
return rc;
/* Cache the xattr object id, valid for the life of the object */
rc = -sa_lookup(obj->oo_sa_hdl, SA_ZPL_XATTR(uos), &obj->oo_xattr, 8);
if (rc == -ENOENT) {
obj->oo_xattr = ZFS_NO_OBJECT;
rc = 0;
} else if (rc) {
osd_object_sa_fini(obj);
}
return rc;
}
/*
* Create the '.zfs' directory.
*/
void
zfsctl_create(zfsvfs_t *zfsvfs)
{
zfsctl_root_t *dot_zfs;
sfs_node_t *snapdir;
vnode_t *rvp;
uint64_t crtime[2];
ASSERT(zfsvfs->z_ctldir == NULL);
snapdir = sfs_alloc_node(sizeof(*snapdir), "snapshot", ZFSCTL_INO_ROOT,
ZFSCTL_INO_SNAPDIR);
dot_zfs = (zfsctl_root_t *)sfs_alloc_node(sizeof(*dot_zfs), ".zfs", 0,
ZFSCTL_INO_ROOT);
dot_zfs->snapdir = snapdir;
VERIFY(VFS_ROOT(zfsvfs->z_vfs, LK_EXCLUSIVE, &rvp) == 0);
VERIFY(0 == sa_lookup(VTOZ(rvp)->z_sa_hdl, SA_ZPL_CRTIME(zfsvfs),
&crtime, sizeof(crtime)));
ZFS_TIME_DECODE(&dot_zfs->cmtime, crtime);
vput(rvp);
zfsvfs->z_ctldir = dot_zfs;
}
/*
* 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);
}
/*
* Look up an entry in a directory.
*
* NOTE: '.' and '..' are handled as special cases because
* no directory entries are actually stored for them. If this is
* the root of a filesystem, then '.zfs' is also treated as a
* special pseudo-directory.
*/
int
zfs_dirlook(znode_t *dzp, char *name, vnode_t **vpp, int flags,
int *deflg, pathname_t *rpnp)
{
zfs_dirlock_t *dl;
znode_t *zp;
int error = 0;
uint64_t parent;
if (name[0] == 0 || (name[0] == '.' && name[1] == 0)) {
vnode_t *vn = ZTOV(dzp);
if (VN_HOLD(vn) == NULL)
error = EAGAIN;
else
*vpp = vn;
} else if (name[0] == '.' && name[1] == '.' && name[2] == 0) {
zfsvfs_t *zfsvfs = dzp->z_zfsvfs;
/*
* If we are a snapshot mounted under .zfs, return
* the vp for the snapshot directory.
*/
if ((error = sa_lookup(dzp->z_sa_hdl,
SA_ZPL_PARENT(zfsvfs), &parent, sizeof (parent))) != 0)
return (error);
if (parent == dzp->z_id && zfsvfs->z_parent != zfsvfs) {
error = zfsctl_root_lookup(zfsvfs->z_parent->z_ctldir,
"snapshot", vpp, NULL, 0, NULL, kcred,
NULL, NULL, NULL);
return (error);
}
rw_enter(&dzp->z_parent_lock, RW_READER);
error = zfs_zget(zfsvfs, parent, &zp);
if (error == 0)
*vpp = ZTOV(zp);
rw_exit(&dzp->z_parent_lock);
} else if (zfs_has_ctldir(dzp) && strcmp(name, ZFS_CTLDIR_NAME) == 0) {
vnode_t *vn = zfsctl_root(dzp);
if (vn)
*vpp = vn;
else
error = EAGAIN;
} else {
int zf;
zf = ZEXISTS | ZSHARED;
if (flags & FIGNORECASE)
zf |= ZCILOOK;
error = zfs_dirent_lock(&dl, dzp, name, &zp, zf, deflg, rpnp);
if (error == 0) {
*vpp = ZTOV(zp);
zfs_dirent_unlock(dl);
dzp->z_zn_prefetch = B_TRUE; /* enable prefetching */
}
rpnp = NULL;
}
if ((flags & FIGNORECASE) && rpnp && !error)
(void) strlcpy(rpnp->pn_buf, name, rpnp->pn_bufsize);
return (error);
}
void
zfs_rmnode(znode_t *zp)
{
zfsvfs_t *zfsvfs = zp->z_zfsvfs;
objset_t *os = zfsvfs->z_os;
znode_t *xzp = NULL;
dmu_tx_t *tx;
uint64_t acl_obj;
uint64_t xattr_obj;
int error;
ASSERT(zp->z_links == 0);
ASSERT(ZTOV(zp)->v_count == 0);
/*
* If this is an attribute directory, purge its contents.
*/
if (ZTOV(zp)->v_type == VDIR && (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);
zfs_znode_free(zp);
return;
}
}
/*
* Free up all the data in the file.
*/
error = dmu_free_long_range(os, zp->z_id, 0, DMU_OBJECT_END);
if (error) {
/*
* Not enough space. Leave the file in the unlinked set.
*/
zfs_znode_dmu_fini(zp);
zfs_znode_free(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);
zfs_znode_free(zp);
goto out;
}
if (xzp) {
ASSERT(error == 0);
mutex_enter(&xzp->z_lock);
xzp->z_unlinked = B_TRUE; /* mark xzp for deletion */
xzp->z_links = 0; /* no more links to it */
VERIFY(0 == sa_update(xzp->z_sa_hdl, SA_ZPL_LINKS(zfsvfs),
&xzp->z_links, sizeof (xzp->z_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)
VN_RELE(ZTOV(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 = dzp->z_zfsvfs;
zfs_dirlock_t *dl;
boolean_t update;
boolean_t exact;
uint64_t zoid;
vnode_t *vp = NULL;
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 (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.
*/
/*
* Decide if exact matches should be requested when performing
* a zap lookup on file systems supporting case-insensitive
* access.
*/
exact =
((zfsvfs->z_case == ZFS_CASE_INSENSITIVE) && (flag & ZCIEXACT)) ||
((zfsvfs->z_case == ZFS_CASE_MIXED) && !(flag & ZCILOOK));
/*
* 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) {
mutex_exit(&dzp->z_lock);
if (!(flag & ZHAVELOCK))
rw_exit(&dzp->z_name_lock);
return (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 (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. He'll either see the old value,
* which is his, 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 ? ENOENT : 0);
} else {
if (update)
vp = dnlc_lookup(ZTOV(dzp), name);
if (vp == DNLC_NO_VNODE) {
VN_RELE(vp);
error = ENOENT;
} else if (vp) {
if (flag & ZNEW) {
zfs_dirent_unlock(dl);
VN_RELE(vp);
return (EEXIST);
}
*dlpp = dl;
*zpp = VTOZ(vp);
return (0);
} else {
error = zfs_match_find(zfsvfs, dzp, name, exact,
update, direntflags, realpnp, &zoid);
}
}
if (error) {
if (error != ENOENT || (flag & ZEXISTS)) {
zfs_dirent_unlock(dl);
return (error);
}
} else {
if (flag & ZNEW) {
zfs_dirent_unlock(dl);
return (EEXIST);
}
error = zfs_zget(zfsvfs, zoid, zpp);
if (error) {
zfs_dirent_unlock(dl);
return (error);
}
if (!(flag & ZXATTR) && update)
dnlc_update(ZTOV(dzp), name, ZTOV(*zpp));
}
*dlpp = dl;
return (0);
}
/*
* Link zp into dl. Can only fail if zp has been unlinked.
*/
int
zfs_make_xattrdir(znode_t *zp, vattr_t *vap, vnode_t **xvpp, cred_t *cr)
{
zfsvfs_t *zfsvfs = zp->z_zfsvfs;
znode_t *xzp;
dmu_tx_t *tx;
int error;
zfs_acl_ids_t acl_ids;
boolean_t fuid_dirtied;
uint64_t parent;
*xvpp = 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 (EDQUOT);
}
top:
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_NOWAIT);
if (error) {
if (error == ERESTART) {
dmu_tx_wait(tx);
dmu_tx_abort(tx);
goto top;
}
zfs_acl_ids_free(&acl_ids);
dmu_tx_abort(tx);
return (error);
}
zfs_mknode(zp, vap, tx, cr, IS_XATTR, &xzp, &acl_ids);
#ifdef HAVE_ZPL
if (fuid_dirtied)
zfs_fuid_sync(zfsvfs, tx);
#endif /* HAVE_ZPL */
#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));
(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);
*xvpp = ZTOV(xzp);
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));
}
int
__osd_xattr_set(const struct lu_env *env, struct osd_object *obj,
const struct lu_buf *buf, const char *name, int fl,
struct osd_thandle *oh)
{
struct osd_device *osd = osd_obj2dev(obj);
dmu_buf_t *xa_zap_db = NULL;
dmu_buf_t *xa_data_db = NULL;
uint64_t xa_data_obj;
sa_handle_t *sa_hdl = NULL;
dmu_tx_t *tx = oh->ot_tx;
uint64_t size;
int rc;
LASSERT(obj->oo_sa_hdl);
if (obj->oo_xattr == ZFS_NO_OBJECT) {
struct lu_attr *la = &osd_oti_get(env)->oti_la;
la->la_valid = LA_MODE;
la->la_mode = S_IFDIR | S_IRUGO | S_IWUSR | S_IXUGO;
rc = __osd_zap_create(env, osd, &xa_zap_db, tx, la,
obj->oo_db->db_object, 0);
if (rc)
return rc;
obj->oo_xattr = xa_zap_db->db_object;
rc = osd_object_sa_update(obj, SA_ZPL_XATTR(osd),
&obj->oo_xattr, 8, oh);
if (rc)
goto out;
}
rc = -zap_lookup(osd->od_os, obj->oo_xattr, name, sizeof(uint64_t), 1,
&xa_data_obj);
if (rc == 0) {
if (fl & LU_XATTR_CREATE) {
rc = -EEXIST;
goto out;
}
/*
* Entry already exists.
* We'll truncate the existing object.
*/
rc = __osd_obj2dbuf(env, osd->od_os, xa_data_obj,
&xa_data_db);
if (rc)
goto out;
rc = -sa_handle_get(osd->od_os, xa_data_obj, NULL,
SA_HDL_PRIVATE, &sa_hdl);
if (rc)
goto out;
rc = -sa_lookup(sa_hdl, SA_ZPL_SIZE(osd), &size, 8);
if (rc)
goto out_sa;
rc = -dmu_free_range(osd->od_os, xa_data_db->db_object,
0, DMU_OBJECT_END, tx);
if (rc)
goto out_sa;
} else if (rc == -ENOENT) {
struct lu_attr *la = &osd_oti_get(env)->oti_la;
/*
* Entry doesn't exist, we need to create a new one and a new
* object to store the value.
*/
if (fl & LU_XATTR_REPLACE) {
/* should be ENOATTR according to the
* man, but that is undefined here */
rc = -ENODATA;
goto out;
}
la->la_valid = LA_MODE;
la->la_mode = S_IFREG | S_IRUGO | S_IWUSR;
rc = __osd_object_create(env, obj, &xa_data_db, tx, la,
obj->oo_xattr);
if (rc)
goto out;
xa_data_obj = xa_data_db->db_object;
rc = -sa_handle_get(osd->od_os, xa_data_obj, NULL,
SA_HDL_PRIVATE, &sa_hdl);
if (rc)
goto out;
rc = -zap_add(osd->od_os, obj->oo_xattr, name, sizeof(uint64_t),
1, &xa_data_obj, tx);
if (rc)
goto out_sa;
} else {
/* There was an error looking up the xattr name */
goto out;
}
/* Finally write the xattr value */
dmu_write(osd->od_os, xa_data_obj, 0, buf->lb_len, buf->lb_buf, tx);
size = buf->lb_len;
rc = -sa_update(sa_hdl, SA_ZPL_SIZE(osd), &size, 8, tx);
out_sa:
sa_handle_destroy(sa_hdl);
out:
if (xa_data_db != NULL)
dmu_buf_rele(xa_data_db, FTAG);
if (xa_zap_db != NULL)
dmu_buf_rele(xa_zap_db, FTAG);
return rc;
}
int
zfs_vget(struct super_block *sb, struct inode **ipp, fid_t *fidp)
{
zfs_sb_t *zsb = sb->s_fs_info;
znode_t *zp;
uint64_t object = 0;
uint64_t fid_gen = 0;
uint64_t gen_mask;
uint64_t zp_gen;
int i, err;
*ipp = NULL;
ZFS_ENTER(zsb);
if (fidp->fid_len == LONG_FID_LEN) {
zfid_long_t *zlfid = (zfid_long_t *)fidp;
uint64_t objsetid = 0;
uint64_t setgen = 0;
for (i = 0; i < sizeof (zlfid->zf_setid); i++)
objsetid |= ((uint64_t)zlfid->zf_setid[i]) << (8 * i);
for (i = 0; i < sizeof (zlfid->zf_setgen); i++)
setgen |= ((uint64_t)zlfid->zf_setgen[i]) << (8 * i);
ZFS_EXIT(zsb);
err = zfsctl_lookup_objset(sb, objsetid, &zsb);
if (err)
return (SET_ERROR(EINVAL));
ZFS_ENTER(zsb);
}
if (fidp->fid_len == SHORT_FID_LEN || fidp->fid_len == LONG_FID_LEN) {
zfid_short_t *zfid = (zfid_short_t *)fidp;
for (i = 0; i < sizeof (zfid->zf_object); i++)
object |= ((uint64_t)zfid->zf_object[i]) << (8 * i);
for (i = 0; i < sizeof (zfid->zf_gen); i++)
fid_gen |= ((uint64_t)zfid->zf_gen[i]) << (8 * i);
} else {
ZFS_EXIT(zsb);
return (SET_ERROR(EINVAL));
}
/* A zero fid_gen means we are in the .zfs control directories */
if (fid_gen == 0 &&
(object == ZFSCTL_INO_ROOT || object == ZFSCTL_INO_SNAPDIR)) {
*ipp = zsb->z_ctldir;
ASSERT(*ipp != NULL);
if (object == ZFSCTL_INO_SNAPDIR) {
VERIFY(zfsctl_root_lookup(*ipp, "snapshot", ipp,
0, kcred, NULL, NULL) == 0);
} else {
igrab(*ipp);
}
ZFS_EXIT(zsb);
return (0);
}
gen_mask = -1ULL >> (64 - 8 * i);
dprintf("getting %llu [%llu mask %llx]\n", object, fid_gen, gen_mask);
if ((err = zfs_zget(zsb, object, &zp))) {
ZFS_EXIT(zsb);
return (err);
}
(void) sa_lookup(zp->z_sa_hdl, SA_ZPL_GEN(zsb), &zp_gen,
sizeof (uint64_t));
zp_gen = zp_gen & gen_mask;
if (zp_gen == 0)
zp_gen = 1;
if (zp->z_unlinked || zp_gen != fid_gen) {
dprintf("znode gen (%llu) != fid gen (%llu)\n", zp_gen,
fid_gen);
iput(ZTOI(zp));
ZFS_EXIT(zsb);
return (SET_ERROR(EINVAL));
}
*ipp = ZTOI(zp);
if (*ipp)
zfs_inode_update(ITOZ(*ipp));
ZFS_EXIT(zsb);
return (0);
}
