* CDDL HEADER END
*/
/*
- * Copyright 2008 Sun Microsystems, Inc. All rights reserved.
- * Use is subject to license terms.
+ * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
*/
+
#include <sys/types.h>
#include <sys/param.h>
#include <sys/time.h>
#include <sys/atomic.h>
#include <sys/zfs_ctldir.h>
#include <sys/zfs_fuid.h>
+#include <sys/sa.h>
+#include <sys/zfs_sa.h>
#include <sys/dnlc.h>
#include <sys/extdirent.h>
* of names after deciding which is the appropriate lookup interface.
*/
static int
-zfs_match_find(zfsvfs_t *zfsvfs, znode_t *dzp, char *name, boolean_t exact,
+zfs_match_find(zfs_sb_t *zsb, znode_t *dzp, char *name, boolean_t exact,
boolean_t update, int *deflags, pathname_t *rpnp, uint64_t *zoid)
{
+ boolean_t conflict = B_FALSE;
int error;
- if (zfsvfs->z_norm) {
+ if (zsb->z_norm) {
matchtype_t mt = MT_FIRST;
- boolean_t conflict = B_FALSE;
size_t bufsz = 0;
char *buf = NULL;
* In the non-mixed case we only expect there would ever
* be one match, but we need to use the normalizing lookup.
*/
- error = zap_lookup_norm(zfsvfs->z_os, dzp->z_id, name, 8, 1,
+ error = zap_lookup_norm(zsb->z_os, dzp->z_id, name, 8, 1,
zoid, mt, buf, bufsz, &conflict);
- if (!error && deflags)
- *deflags = conflict ? ED_CASE_CONFLICT : 0;
} else {
- error = zap_lookup(zfsvfs->z_os, dzp->z_id, name, 8, 1, zoid);
+ error = zap_lookup(zsb->z_os, dzp->z_id, name, 8, 1, zoid);
}
+
+ /*
+ * Allow multiple entries provided the first entry is
+ * the object id. Non-zpl consumers may safely make
+ * use of the additional space.
+ *
+ * XXX: This should be a feature flag for compatibility
+ */
+ if (error == EOVERFLOW)
+ error = 0;
+
+ if (zsb->z_norm && !error && deflags)
+ *deflags = conflict ? ED_CASE_CONFLICT : 0;
+
*zoid = ZFS_DIRENT_OBJ(*zoid);
+#ifdef HAVE_DNLC
if (error == ENOENT && update)
- dnlc_update(ZTOV(dzp), name, DNLC_NO_VNODE);
+ dnlc_update(ZTOI(dzp), name, DNLC_NO_VNODE);
+#endif /* HAVE_DNLC */
return (error);
}
* 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)
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_sb_t *zsb = ZTOZSB(dzp);
zfs_dirlock_t *dl;
boolean_t update;
boolean_t exact;
uint64_t zoid;
+#ifdef HAVE_DNLC
vnode_t *vp = NULL;
+#endif /* HAVE_DNLC */
int error = 0;
int cmpflags;
/*
* 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)
+ 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
+ * zsb->z_case and zsb->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.
*
* access.
*/
exact =
- ((zfsvfs->z_case == ZFS_CASE_INSENSITIVE) && (flag & ZCIEXACT)) ||
- ((zfsvfs->z_case == ZFS_CASE_MIXED) && !(flag & ZCILOOK));
+ ((zsb->z_case == ZFS_CASE_INSENSITIVE) && (flag & ZCIEXACT)) ||
+ ((zsb->z_case == ZFS_CASE_MIXED) && !(flag & ZCILOOK));
/*
* Only look in or update the DNLC if we are looking for the
* 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));
+ update = !zsb->z_norm ||
+ ((zsb->z_case == ZFS_CASE_MIXED) &&
+ !(zsb->z_norm & ~U8_TEXTPREP_TOUPPER) && !(flag & ZCILOOK));
/*
* ZRENAMING indicates we are in a situation where we should
if (flag & ZRENAMING)
cmpflags = 0;
else
- cmpflags = zfsvfs->z_norm;
+ cmpflags = zsb->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.
*/
- rw_enter(&dzp->z_name_lock, RW_READER);
+ 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);
- rw_exit(&dzp->z_name_lock);
+ if (!(flag & ZHAVELOCK))
+ rw_exit(&dzp->z_name_lock);
return (ENOENT);
}
for (dl = dzp->z_dirlocks; dl != NULL; dl = dl->dl_next) {
}
if (error != 0) {
mutex_exit(&dzp->z_lock);
- rw_exit(&dzp->z_name_lock);
+ if (!(flag & ZHAVELOCK))
+ rw_exit(&dzp->z_name_lock);
return (ENOENT);
}
if (dl == NULL) {
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;
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
* See if there's an object by this name; if so, put a hold on it.
*/
if (flag & ZXATTR) {
- zoid = dzp->z_phys->zp_xattr;
- error = (zoid == 0 ? ENOENT : 0);
+ error = sa_lookup(dzp->z_sa_hdl, SA_ZPL_XATTR(zsb), &zoid,
+ sizeof (zoid));
+ if (error == 0)
+ error = (zoid == 0 ? ENOENT : 0);
} else {
+#ifdef HAVE_DNLC
if (update)
- vp = dnlc_lookup(ZTOV(dzp), name);
+ vp = dnlc_lookup(ZTOI(dzp), name);
if (vp == DNLC_NO_VNODE) {
- VN_RELE(vp);
+ iput(vp);
error = ENOENT;
} else if (vp) {
if (flag & ZNEW) {
zfs_dirent_unlock(dl);
- VN_RELE(vp);
+ iput(vp);
return (EEXIST);
}
*dlpp = dl;
*zpp = VTOZ(vp);
return (0);
} else {
- error = zfs_match_find(zfsvfs, dzp, name, exact,
+ error = zfs_match_find(zsb, dzp, name, exact,
update, direntflags, realpnp, &zoid);
}
+#else
+ error = zfs_match_find(zsb, dzp, name, exact,
+ update, direntflags, realpnp, &zoid);
+#endif /* HAVE_DNLC */
}
if (error) {
if (error != ENOENT || (flag & ZEXISTS)) {
zfs_dirent_unlock(dl);
return (EEXIST);
}
- error = zfs_zget(zfsvfs, zoid, zpp);
+ error = zfs_zget(zsb, zoid, zpp);
if (error) {
zfs_dirent_unlock(dl);
return (error);
}
+#ifdef HAVE_DNLC
if (!(flag & ZXATTR) && update)
- dnlc_update(ZTOV(dzp), name, ZTOV(*zpp));
+ dnlc_update(ZTOI(dzp), name, ZTOI(*zpp));
+#endif /* HAVE_DNLC */
}
*dlpp = dl;
zfs_dirlock_t **prev_dl, *cur_dl;
mutex_enter(&dzp->z_lock);
- rw_exit(&dzp->z_name_lock);
+
+ if (!dl->dl_namelock)
+ rw_exit(&dzp->z_name_lock);
+
if (dl->dl_sharecnt > 1) {
dl->dl_sharecnt--;
mutex_exit(&dzp->z_lock);
* special pseudo-directory.
*/
int
-zfs_dirlook(znode_t *dzp, char *name, vnode_t **vpp, int flags,
+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)) {
- *vpp = ZTOV(dzp);
- VN_HOLD(*vpp);
+ *ipp = ZTOI(dzp);
+ igrab(*ipp);
} else if (name[0] == '.' && name[1] == '.' && name[2] == 0) {
- zfsvfs_t *zfsvfs = dzp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(dzp);
+
/*
* If we are a snapshot mounted under .zfs, return
- * the vp for the snapshot directory.
+ * the inode pointer for the snapshot directory.
*/
- if (dzp->z_phys->zp_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);
+ if ((error = sa_lookup(dzp->z_sa_hdl,
+ SA_ZPL_PARENT(zsb), &parent, sizeof (parent))) != 0)
+ return (error);
+
+ if (parent == dzp->z_id && zsb->z_parent != zsb) {
+ error = zfsctl_root_lookup(zsb->z_parent->z_ctldir,
+ "snapshot", ipp, 0, kcred, NULL, NULL);
return (error);
}
rw_enter(&dzp->z_parent_lock, RW_READER);
- error = zfs_zget(zfsvfs, dzp->z_phys->zp_parent, &zp);
+ error = zfs_zget(zsb, parent, &zp);
if (error == 0)
- *vpp = ZTOV(zp);
+ *ipp = ZTOI(zp);
rw_exit(&dzp->z_parent_lock);
} else if (zfs_has_ctldir(dzp) && strcmp(name, ZFS_CTLDIR_NAME) == 0) {
- *vpp = zfsctl_root(dzp);
+ *ipp = zfsctl_root(dzp);
} else {
int zf;
error = zfs_dirent_lock(&dl, dzp, name, &zp, zf, deflg, rpnp);
if (error == 0) {
- *vpp = ZTOV(zp);
+ *ipp = ZTOI(zp);
zfs_dirent_unlock(dl);
dzp->z_zn_prefetch = B_TRUE; /* enable prefetching */
}
void
zfs_unlinked_add(znode_t *zp, dmu_tx_t *tx)
{
- zfsvfs_t *zfsvfs = zp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(zp);
ASSERT(zp->z_unlinked);
- ASSERT3U(zp->z_phys->zp_links, ==, 0);
+ ASSERT(zp->z_links == 0);
VERIFY3U(0, ==,
- zap_add_int(zfsvfs->z_os, zfsvfs->z_unlinkedobj, zp->z_id, tx));
-}
-
-/*
- * Clean up any znodes that had no links when we either crashed or
- * (force) umounted the file system.
- */
-void
-zfs_unlinked_drain(zfsvfs_t *zfsvfs)
-{
- zap_cursor_t zc;
- zap_attribute_t zap;
- dmu_object_info_t doi;
- znode_t *zp;
- int error;
-
- /*
- * Interate over the contents of the unlinked set.
- */
- for (zap_cursor_init(&zc, zfsvfs->z_os, zfsvfs->z_unlinkedobj);
- zap_cursor_retrieve(&zc, &zap) == 0;
- zap_cursor_advance(&zc)) {
-
- /*
- * See what kind of object we have in list
- */
-
- error = dmu_object_info(zfsvfs->z_os,
- zap.za_first_integer, &doi);
- if (error != 0)
- continue;
-
- ASSERT((doi.doi_type == DMU_OT_PLAIN_FILE_CONTENTS) ||
- (doi.doi_type == DMU_OT_DIRECTORY_CONTENTS));
- /*
- * We need to re-mark these list entries for deletion,
- * so we pull them back into core and set zp->z_unlinked.
- */
- error = zfs_zget(zfsvfs, zap.za_first_integer, &zp);
-
- /*
- * We may pick up znodes that are already marked for deletion.
- * This could happen during the purge of an extended attribute
- * directory. All we need to do is skip over them, since they
- * are already in the system marked z_unlinked.
- */
- if (error != 0)
- continue;
-
- zp->z_unlinked = B_TRUE;
- VN_RELE(ZTOV(zp));
- }
- zap_cursor_fini(&zc);
+ zap_add_int(zsb->z_os, zsb->z_unlinkedobj, zp->z_id, tx));
}
/*
zap_attribute_t zap;
znode_t *xzp;
dmu_tx_t *tx;
- zfsvfs_t *zfsvfs = dzp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(dzp);
zfs_dirlock_t dl;
int skipped = 0;
int error;
- for (zap_cursor_init(&zc, zfsvfs->z_os, dzp->z_id);
+ for (zap_cursor_init(&zc, zsb->z_os, dzp->z_id);
(error = zap_cursor_retrieve(&zc, &zap)) == 0;
zap_cursor_advance(&zc)) {
- error = zfs_zget(zfsvfs,
+ error = zfs_zget(zsb,
ZFS_DIRENT_OBJ(zap.za_first_integer), &xzp);
if (error) {
skipped += 1;
continue;
}
- ASSERT((ZTOV(xzp)->v_type == VREG) ||
- (ZTOV(xzp)->v_type == VLNK));
+ ASSERT(S_ISREG(ZTOI(xzp)->i_mode)||S_ISLNK(ZTOI(xzp)->i_mode));
- tx = dmu_tx_create(zfsvfs->z_os);
- dmu_tx_hold_bonus(tx, dzp->z_id);
+ tx = dmu_tx_create(zsb->z_os);
+ dmu_tx_hold_sa(tx, dzp->z_sa_hdl, B_FALSE);
dmu_tx_hold_zap(tx, dzp->z_id, FALSE, zap.za_name);
- dmu_tx_hold_bonus(tx, xzp->z_id);
- dmu_tx_hold_zap(tx, zfsvfs->z_unlinkedobj, FALSE, NULL);
+ dmu_tx_hold_sa(tx, xzp->z_sa_hdl, B_FALSE);
+ dmu_tx_hold_zap(tx, zsb->z_unlinkedobj, FALSE, NULL);
+ /* Is this really needed ? */
+ zfs_sa_upgrade_txholds(tx, xzp);
error = dmu_tx_assign(tx, TXG_WAIT);
if (error) {
dmu_tx_abort(tx);
- VN_RELE(ZTOV(xzp));
+ iput(ZTOI(xzp));
skipped += 1;
continue;
}
skipped += 1;
dmu_tx_commit(tx);
- VN_RELE(ZTOV(xzp));
+ iput(ZTOI(xzp));
}
zap_cursor_fini(&zc);
if (error != ENOENT)
return (skipped);
}
+/*
+ * Clean up any znodes that had no links when we either crashed or
+ * (force) umounted the file system.
+ */
+void
+zfs_unlinked_drain(zfs_sb_t *zsb)
+{
+ zap_cursor_t zc;
+ zap_attribute_t zap;
+ dmu_object_info_t doi;
+ znode_t *zp;
+ int error;
+
+ /*
+ * Iterate over the contents of the unlinked set.
+ */
+ for (zap_cursor_init(&zc, zsb->z_os, zsb->z_unlinkedobj);
+ zap_cursor_retrieve(&zc, &zap) == 0;
+ zap_cursor_advance(&zc)) {
+
+ /*
+ * See what kind of object we have in list
+ */
+
+ error = dmu_object_info(zsb->z_os, zap.za_first_integer, &doi);
+ if (error != 0)
+ continue;
+
+ ASSERT((doi.doi_type == DMU_OT_PLAIN_FILE_CONTENTS) ||
+ (doi.doi_type == DMU_OT_DIRECTORY_CONTENTS));
+ /*
+ * We need to re-mark these list entries for deletion,
+ * so we pull them back into core and set zp->z_unlinked.
+ */
+ error = zfs_zget(zsb, zap.za_first_integer, &zp);
+
+ /*
+ * We may pick up znodes that are already marked for deletion.
+ * This could happen during the purge of an extended attribute
+ * directory. All we need to do is skip over them, since they
+ * are already in the system marked z_unlinked.
+ */
+ if (error != 0)
+ continue;
+
+ zp->z_unlinked = B_TRUE;
+
+ /*
+ * If this is an attribute directory, purge its contents.
+ */
+ if (S_ISDIR(ZTOI(zp)->i_mode) && (zp->z_pflags & ZFS_XATTR)) {
+ /*
+ * We don't need to check the return value of
+ * zfs_purgedir here, because zfs_rmnode will just
+ * return this xattr directory to the unlinked set
+ * until all of its xattrs are gone.
+ */
+ (void) zfs_purgedir(zp);
+ }
+
+ iput(ZTOI(zp));
+ }
+ zap_cursor_fini(&zc);
+}
+
void
zfs_rmnode(znode_t *zp)
{
- zfsvfs_t *zfsvfs = zp->z_zfsvfs;
- objset_t *os = zfsvfs->z_os;
+ zfs_sb_t *zsb = ZTOZSB(zp);
+ objset_t *os = zsb->z_os;
znode_t *xzp = NULL;
dmu_tx_t *tx;
uint64_t acl_obj;
+ uint64_t xattr_obj;
+ uint64_t count;
int error;
- ASSERT(ZTOV(zp)->v_count == 0);
- ASSERT(zp->z_phys->zp_links == 0);
-
- /*
- * If this is a ZIL replay then leave the object in the unlinked set.
- * Otherwise we can get a deadlock, because the delete can be
- * quite large and span multiple tx's and txgs, but each replay
- * creates a tx to atomically run the replay function and mark the
- * replay record as complete. We deadlock trying to start a tx in
- * a new txg to further the deletion but can't because the replay
- * tx hasn't finished.
- *
- * We actually delete the object if we get a failure to create an
- * object in zil_replay_log_record(), or after calling zil_replay().
- */
- if (zfsvfs->z_assign >= TXG_INITIAL) {
- zfs_znode_dmu_fini(zp);
- zfs_znode_free(zp);
- return;
- }
+ ASSERT(zp->z_links == 0);
+ ASSERT(atomic_read(&ZTOI(zp)->i_count) == 0);
/*
* If this is an attribute directory, purge its contents.
*/
- if (ZTOV(zp)->v_type == VDIR && (zp->z_phys->zp_flags & ZFS_XATTR)) {
- if (zfs_purgedir(zp) != 0) {
+ if (S_ISDIR(ZTOI(zp)->i_mode) && (zp->z_pflags & ZFS_XATTR)) {
+ error = zap_count(os, zp->z_id, &count);
+ if (error) {
+ zfs_znode_dmu_fini(zp);
+ return;
+ }
+
+ if (count > 0) {
+ taskq_t *taskq;
+
/*
- * Not enough space to delete some xattrs.
- * Leave it in the unlinked set.
+ * There are still directory entries in this xattr
+ * directory. Let zfs_unlinked_drain() deal with
+ * them to avoid deadlocking this process in the
+ * zfs_purgedir()->zfs_zget()->ilookup() callpath
+ * on the xattr inode's I_FREEING bit.
*/
+ taskq = dsl_pool_iput_taskq(dmu_objset_pool(os));
+ taskq_dispatch(taskq, (task_func_t *)
+ zfs_unlinked_drain, zsb, TQ_SLEEP);
+
zfs_znode_dmu_fini(zp);
- zfs_znode_free(zp);
return;
}
}
* 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.
*/
- if (zp->z_phys->zp_xattr) {
- error = zfs_zget(zfsvfs, zp->z_phys->zp_xattr, &xzp);
+ error = sa_lookup(zp->z_sa_hdl, SA_ZPL_XATTR(zsb),
+ &xattr_obj, sizeof (xattr_obj));
+ if (error == 0 && xattr_obj) {
+ error = zfs_zget(zsb, xattr_obj, &xzp);
ASSERT(error == 0);
}
- acl_obj = zp->z_phys->zp_acl.z_acl_extern_obj;
+ 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);
+ dmu_tx_hold_zap(tx, zsb->z_unlinkedobj, FALSE, NULL);
if (xzp) {
- dmu_tx_hold_bonus(tx, xzp->z_id);
- dmu_tx_hold_zap(tx, zfsvfs->z_unlinkedobj, TRUE, NULL);
+ dmu_tx_hold_zap(tx, zsb->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) {
/*
*/
dmu_tx_abort(tx);
zfs_znode_dmu_fini(zp);
- zfs_znode_free(zp);
goto out;
}
if (xzp) {
- dmu_buf_will_dirty(xzp->z_dbuf, tx);
+ ASSERT(error == 0);
mutex_enter(&xzp->z_lock);
xzp->z_unlinked = B_TRUE; /* mark xzp for deletion */
- xzp->z_phys->zp_links = 0; /* no more links to it */
+ xzp->z_links = 0; /* no more links to it */
+ VERIFY(0 == sa_update(xzp->z_sa_hdl, SA_ZPL_LINKS(zsb),
+ &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));
+ zap_remove_int(zsb->z_os, zsb->z_unlinkedobj, zp->z_id, tx));
zfs_znode_delete(zp, tx);
dmu_tx_commit(tx);
out:
if (xzp)
- VN_RELE(ZTOV(xzp));
+ iput(ZTOI(xzp));
}
static uint64_t
-zfs_dirent(znode_t *zp)
+zfs_dirent(znode_t *zp, uint64_t mode)
{
uint64_t de = zp->z_id;
- if (zp->z_zfsvfs->z_version >= ZPL_VERSION_DIRENT_TYPE)
- de |= IFTODT((zp)->z_phys->zp_mode) << 60;
+
+ if (ZTOZSB(zp)->z_version >= ZPL_VERSION_DIRENT_TYPE)
+ de |= IFTODT(mode) << 60;
return (de);
}
zfs_link_create(zfs_dirlock_t *dl, znode_t *zp, dmu_tx_t *tx, int flag)
{
znode_t *dzp = dl->dl_dzp;
- vnode_t *vp = ZTOV(zp);
+ zfs_sb_t *zsb = ZTOZSB(zp);
uint64_t value;
- int zp_is_dir = (vp->v_type == VDIR);
+ 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;
- dmu_buf_will_dirty(zp->z_dbuf, tx);
mutex_enter(&zp->z_lock);
if (!(flag & ZRENAMING)) {
mutex_exit(&zp->z_lock);
return (ENOENT);
}
- zp->z_phys->zp_links++;
+ zp->z_links++;
+ SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zsb), NULL,
+ &zp->z_links, sizeof (zp->z_links));
+
}
- zp->z_phys->zp_parent = dzp->z_id; /* dzp is now zp's parent */
+ 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);
- if (!(flag & ZNEW))
- zfs_time_stamper_locked(zp, STATE_CHANGED, tx);
mutex_exit(&zp->z_lock);
- dmu_buf_will_dirty(dzp->z_dbuf, tx);
mutex_enter(&dzp->z_lock);
- dzp->z_phys->zp_size++; /* one dirent added */
- dzp->z_phys->zp_links += zp_is_dir; /* ".." link from zp */
- zfs_time_stamper_locked(dzp, CONTENT_MODIFIED, tx);
+ 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);
- error = zap_add(zp->z_zfsvfs->z_os, dzp->z_id, dl->dl_name,
+ 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);
- dnlc_update(ZTOV(dzp), dl->dl_name, vp);
-
return (0);
}
+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);
+}
+
/*
- * 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 (EEXIST).
+ * 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.
boolean_t *unlinkedp)
{
znode_t *dzp = dl->dl_dzp;
- vnode_t *vp = ZTOV(zp);
- int zp_is_dir = (vp->v_type == VDIR);
+ zfs_sb_t *zsb = 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];
+ int count = 0;
int error;
- dnlc_remove(ZTOV(dzp), dl->dl_name);
+#ifdef HAVE_DNLC
+ dnlc_remove(ZTOI(dzp), dl->dl_name);
+#endif /* HAVE_DNLC */
if (!(flag & ZRENAMING)) {
- dmu_buf_will_dirty(zp->z_dbuf, tx);
-
- if (vn_vfswlock(vp)) /* prevent new mounts on zp */
- return (EBUSY);
+ mutex_enter(&zp->z_lock);
- if (vn_ismntpt(vp)) { /* don't remove mount point */
- vn_vfsunlock(vp);
- return (EBUSY);
+ if (zp_is_dir && !zfs_dirempty(zp)) {
+ mutex_exit(&zp->z_lock);
+ return (ENOTEMPTY);
}
- mutex_enter(&zp->z_lock);
- if (zp_is_dir && !zfs_dirempty(zp)) { /* dir not empty */
+ /*
+ * 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);
- vn_vfsunlock(vp);
- return (EEXIST);
+ return (error);
}
- if (zp->z_phys->zp_links <= zp_is_dir) {
- zfs_panic_recover("zfs: link count on %s is %u, "
- "should be at least %u",
- zp->z_vnode->v_path ? zp->z_vnode->v_path :
- "<unknown>", (int)zp->z_phys->zp_links,
- zp_is_dir + 1);
- zp->z_phys->zp_links = zp_is_dir + 1;
+
+ if (zp->z_links <= zp_is_dir) {
+ zfs_panic_recover("zfs: link count on %lu is %u, "
+ "should be at least %u", zp->z_id,
+ (int)zp->z_links, zp_is_dir + 1);
+ zp->z_links = zp_is_dir + 1;
}
- if (--zp->z_phys->zp_links == zp_is_dir) {
+ if (--zp->z_links == zp_is_dir) {
zp->z_unlinked = B_TRUE;
- zp->z_phys->zp_links = 0;
+ zp->z_links = 0;
unlinked = B_TRUE;
} else {
- zfs_time_stamper_locked(zp, STATE_CHANGED, tx);
+ 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, &zp->z_pflags, sizeof (zp->z_pflags));
+ zfs_tstamp_update_setup(zp, STATE_CHANGED, mtime, ctime,
+ B_TRUE);
}
+ SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_LINKS(zsb),
+ NULL, &zp->z_links, sizeof (zp->z_links));
+ error = sa_bulk_update(zp->z_sa_hdl, bulk, count, tx);
+ count = 0;
+ ASSERT(error == 0);
mutex_exit(&zp->z_lock);
- vn_vfsunlock(vp);
+ } else {
+ error = zfs_dropname(dl, zp, dzp, tx, flag);
+ if (error != 0)
+ return (error);
}
- dmu_buf_will_dirty(dzp->z_dbuf, tx);
mutex_enter(&dzp->z_lock);
- dzp->z_phys->zp_size--; /* one dirent removed */
- dzp->z_phys->zp_links -= zp_is_dir; /* ".." link from zp */
- zfs_time_stamper_locked(dzp, CONTENT_MODIFIED, tx);
- mutex_exit(&dzp->z_lock);
-
- if (zp->z_zfsvfs->z_norm) {
- if (((zp->z_zfsvfs->z_case == ZFS_CASE_INSENSITIVE) &&
- (flag & ZCIEXACT)) ||
- ((zp->z_zfsvfs->z_case == ZFS_CASE_MIXED) &&
- !(flag & ZCILOOK)))
- error = zap_remove_norm(zp->z_zfsvfs->z_os,
- dzp->z_id, dl->dl_name, MT_EXACT, tx);
- else
- error = zap_remove_norm(zp->z_zfsvfs->z_os,
- dzp->z_id, dl->dl_name, MT_FIRST, tx);
- } else {
- error = zap_remove(zp->z_zfsvfs->z_os,
- dzp->z_id, dl->dl_name, tx);
- }
+ dzp->z_size--; /* one dirent removed */
+ dzp->z_links -= zp_is_dir; /* ".." link from zp */
+ 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_SIZE(zsb),
+ NULL, &dzp->z_size, sizeof (dzp->z_size));
+ SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_CTIME(zsb),
+ NULL, ctime, sizeof (ctime));
+ SA_ADD_BULK_ATTR(bulk, count, SA_ZPL_MTIME(zsb),
+ NULL, mtime, sizeof (mtime));
+ 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);
if (unlinkedp != NULL)
*unlinkedp = unlinked;
boolean_t
zfs_dirempty(znode_t *dzp)
{
- return (dzp->z_phys->zp_size == 2 && dzp->z_dirlocks == 0);
+ return (dzp->z_size == 2 && dzp->z_dirlocks == 0);
}
int
-zfs_make_xattrdir(znode_t *zp, vattr_t *vap, vnode_t **xvpp, cred_t *cr)
+zfs_make_xattrdir(znode_t *zp, vattr_t *vap, struct inode **xipp, cred_t *cr)
{
- zfsvfs_t *zfsvfs = zp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(zp);
znode_t *xzp;
dmu_tx_t *tx;
int error;
- zfs_fuid_info_t *fuidp = NULL;
+ zfs_acl_ids_t acl_ids;
+ boolean_t fuid_dirtied;
+#ifdef DEBUG
+ uint64_t parent;
+#endif
- *xvpp = NULL;
+ *xipp = NULL;
- if (error = zfs_zaccess(zp, ACE_WRITE_NAMED_ATTRS, 0, B_FALSE, cr))
+ if ((error = zfs_zaccess(zp, ACE_WRITE_NAMED_ATTRS, 0, B_FALSE, cr)))
return (error);
- tx = dmu_tx_create(zfsvfs->z_os);
- dmu_tx_hold_bonus(tx, zp->z_id);
- dmu_tx_hold_zap(tx, DMU_NEW_OBJECT, FALSE, NULL);
- if (IS_EPHEMERAL(crgetuid(cr)) || IS_EPHEMERAL(crgetgid(cr))) {
- if (zfsvfs->z_fuid_obj == 0) {
- dmu_tx_hold_bonus(tx, DMU_NEW_OBJECT);
- dmu_tx_hold_write(tx, DMU_NEW_OBJECT, 0,
- FUID_SIZE_ESTIMATE(zfsvfs));
- dmu_tx_hold_zap(tx, MASTER_NODE_OBJ, FALSE, NULL);
- } else {
- dmu_tx_hold_bonus(tx, zfsvfs->z_fuid_obj);
- dmu_tx_hold_write(tx, zfsvfs->z_fuid_obj, 0,
- FUID_SIZE_ESTIMATE(zfsvfs));
- }
+ if ((error = zfs_acl_ids_create(zp, IS_XATTR, vap, cr, NULL,
+ &acl_ids)) != 0)
+ return (error);
+ if (zfs_acl_ids_overquota(zsb, &acl_ids)) {
+ zfs_acl_ids_free(&acl_ids);
+ return (EDQUOT);
}
- error = dmu_tx_assign(tx, zfsvfs->z_assign);
+
+top:
+ tx = dmu_tx_create(zsb->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 = zsb->z_fuid_dirty;
+ if (fuid_dirtied)
+ zfs_fuid_txhold(zsb, tx);
+ error = dmu_tx_assign(tx, TXG_NOWAIT);
if (error) {
- if (error == ERESTART && zfsvfs->z_assign == TXG_NOWAIT)
+ 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, 0, NULL, &fuidp);
- ASSERT(xzp->z_phys->zp_parent == zp->z_id);
- dmu_buf_will_dirty(zp->z_dbuf, tx);
- zp->z_phys->zp_xattr = xzp->z_id;
-
- (void) zfs_log_create(zfsvfs->z_log, tx, TX_MKXATTR, zp,
- xzp, "", NULL, fuidp, vap);
- if (fuidp)
- zfs_fuid_info_free(fuidp);
+ zfs_mknode(zp, vap, tx, cr, IS_XATTR, &xzp, &acl_ids);
+
+ if (fuid_dirtied)
+ zfs_fuid_sync(zsb, tx);
+
+#ifdef DEBUG
+ error = sa_lookup(xzp->z_sa_hdl, SA_ZPL_PARENT(zsb),
+ &parent, sizeof (parent));
+ ASSERT(error == 0 && parent == zp->z_id);
+#endif
+
+ VERIFY(0 == sa_update(zp->z_sa_hdl, SA_ZPL_XATTR(zsb), &xzp->z_id,
+ sizeof (xzp->z_id), tx));
+
+ (void) zfs_log_create(zsb->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);
+ *xipp = ZTOI(xzp);
return (0);
}
* cr - credentials of caller
* flags - flags from the VOP_LOOKUP call
*
- * OUT: xzpp - pointer to extended attribute znode
+ * OUT: xipp - pointer to extended attribute znode
*
* RETURN: 0 on success
* error number on failure
*/
int
-zfs_get_xattrdir(znode_t *zp, vnode_t **xvpp, cred_t *cr, int flags)
+zfs_get_xattrdir(znode_t *zp, struct inode **xipp, cred_t *cr, int flags)
{
- zfsvfs_t *zfsvfs = zp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(zp);
znode_t *xzp;
zfs_dirlock_t *dl;
vattr_t va;
return (error);
if (xzp != NULL) {
- *xvpp = ZTOV(xzp);
+ *xipp = ZTOI(xzp);
zfs_dirent_unlock(dl);
return (0);
}
- ASSERT(zp->z_phys->zp_xattr == 0);
-
if (!(flags & CREATE_XATTR_DIR)) {
zfs_dirent_unlock(dl);
return (ENOENT);
}
- if (zfsvfs->z_vfs->vfs_flag & VFS_RDONLY) {
+ if (zfs_is_readonly(zsb)) {
zfs_dirent_unlock(dl);
return (EROFS);
}
* Once in a directory the ability to read/write attributes
* is controlled by the permissions on the attribute file.
*/
- va.va_mask = AT_TYPE | AT_MODE | AT_UID | AT_GID;
- va.va_type = VDIR;
+ va.va_mask = ATTR_MODE | ATTR_UID | ATTR_GID;
va.va_mode = S_IFDIR | S_ISVTX | 0777;
zfs_fuid_map_ids(zp, cr, &va.va_uid, &va.va_gid);
- error = zfs_make_xattrdir(zp, &va, xvpp, cr);
+ va.va_dentry = NULL;
+ error = zfs_make_xattrdir(zp, &va, xipp, cr);
zfs_dirent_unlock(dl);
- if (error == ERESTART && zfsvfs->z_assign == TXG_NOWAIT) {
+ if (error == ERESTART) {
/* NB: we already did dmu_tx_wait() if necessary */
goto top;
}
int
zfs_sticky_remove_access(znode_t *zdp, znode_t *zp, cred_t *cr)
{
- uid_t uid;
+ uid_t uid;
uid_t downer;
uid_t fowner;
- zfsvfs_t *zfsvfs = zdp->z_zfsvfs;
+ zfs_sb_t *zsb = ZTOZSB(zdp);
- if (zdp->z_zfsvfs->z_assign >= TXG_INITIAL) /* ZIL replay */
+ if (zsb->z_replay)
return (0);
- if ((zdp->z_phys->zp_mode & S_ISVTX) == 0)
+ if ((zdp->z_mode & S_ISVTX) == 0)
return (0);
- downer = zfs_fuid_map_id(zfsvfs, zdp->z_phys->zp_uid, cr, ZFS_OWNER);
- fowner = zfs_fuid_map_id(zfsvfs, zp->z_phys->zp_uid, cr, ZFS_OWNER);
+ 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 ||
- (ZTOV(zp)->v_type == VREG &&
+ (S_ISDIR(ZTOI(zp)->i_mode) &&
zfs_zaccess(zp, ACE_WRITE_DATA, 0, B_FALSE, cr) == 0))
return (0);
else
git://git.camperquake.de / zfs.git / blobdiff