/*
* xfs_rename
*/
int
xfs_rename(
bhv_desc_t *src_dir_bdp,
vname_t *src_vname,
vnode_t *target_dir_vp,
vname_t *target_vname,
cred_t *credp)
{
xfs_trans_t *tp;
xfs_inode_t *src_dp, *target_dp, *src_ip, *target_ip;
xfs_mount_t *mp;
int new_parent; /* moving to a new dir */
int src_is_directory; /* src_name is a directory */
int error;
xfs_bmap_free_t free_list;
xfs_fsblock_t first_block;
int cancel_flags;
int committed;
xfs_inode_t *inodes[4];
int target_ip_dropped = 0; /* dropped target_ip link? */
vnode_t *src_dir_vp;
bhv_desc_t *target_dir_bdp;
int spaceres;
int target_link_zero = 0;
int num_inodes;
char *src_name = VNAME(src_vname);
char *target_name = VNAME(target_vname);
int src_namelen = VNAMELEN(src_vname);
int target_namelen = VNAMELEN(target_vname);
src_dir_vp = BHV_TO_VNODE(src_dir_bdp);
vn_trace_entry(src_dir_vp, "xfs_rename", (inst_t *)__return_address);
vn_trace_entry(target_dir_vp, "xfs_rename", (inst_t *)__return_address);
/*
* Find the XFS behavior descriptor for the target directory
* vnode since it was not handed to us.
*/
target_dir_bdp = vn_bhv_lookup_unlocked(VN_BHV_HEAD(target_dir_vp),
&xfs_vnodeops);
if (target_dir_bdp == NULL) {
return XFS_ERROR(EXDEV);
}
src_dp = XFS_BHVTOI(src_dir_bdp);
target_dp = XFS_BHVTOI(target_dir_bdp);
mp = src_dp->i_mount;
if (DM_EVENT_ENABLED(src_dir_vp->v_vfsp, src_dp, DM_EVENT_RENAME) ||
DM_EVENT_ENABLED(target_dir_vp->v_vfsp,
target_dp, DM_EVENT_RENAME)) {
error = XFS_SEND_NAMESP(mp, DM_EVENT_RENAME,
src_dir_vp, DM_RIGHT_NULL,
target_dir_vp, DM_RIGHT_NULL,
src_name, target_name,
0, 0, 0);
if (error) {
return error;
}
}
/* Return through std_return after this point. */
/*
* Lock all the participating inodes. Depending upon whether
* the target_name exists in the target directory, and
* whether the target directory is the same as the source
* directory, we can lock from 2 to 4 inodes.
* xfs_lock_for_rename() will return ENOENT if src_name
* does not exist in the source directory.
*/
tp = NULL;
error = xfs_lock_for_rename(src_dp, target_dp, src_vname,
target_vname, &src_ip, &target_ip, inodes,
&num_inodes);
if (error) {
/*
* We have nothing locked, no inode references, and
* no transaction, so just get out.
*/
goto std_return;
}
ASSERT(src_ip != NULL);
if ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
/*
* Check for link count overflow on target_dp
*/
if (target_ip == NULL && (src_dp != target_dp) &&
target_dp->i_d.di_nlink >= XFS_MAXLINK) {
error = XFS_ERROR(EMLINK);
xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED);
goto rele_return;
}
}
new_parent = (src_dp != target_dp);
src_is_directory = ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR);
/*
* Drop the locks on our inodes so that we can start the transaction.
*/
xfs_rename_unlock4(inodes, XFS_ILOCK_SHARED);
XFS_BMAP_INIT(&free_list, &first_block);
tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME);
cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
spaceres = XFS_RENAME_SPACE_RES(mp, target_namelen);
error = xfs_trans_reserve(tp, spaceres, XFS_RENAME_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT);
if (error == ENOSPC) {
spaceres = 0;
error = xfs_trans_reserve(tp, 0, XFS_RENAME_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT);
}
if (error) {
xfs_trans_cancel(tp, 0);
goto rele_return;
}
/*
* Attach the dquots to the inodes
*/
if ((error = XFS_QM_DQVOPRENAME(mp, inodes))) {
xfs_trans_cancel(tp, cancel_flags);
goto rele_return;
}
/*
* Reacquire the inode locks we dropped above.
*/
xfs_lock_inodes(inodes, num_inodes, 0, XFS_ILOCK_EXCL);
/*
* Join all the inodes to the transaction. From this point on,
* we can rely on either trans_commit or trans_cancel to unlock
* them. Note that we need to add a vnode reference to the
* directories since trans_commit & trans_cancel will decrement
* them when they unlock the inodes. Also, we need to be careful
* not to add an inode to the transaction more than once.
*/
VN_HOLD(src_dir_vp);
xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL);
if (new_parent) {
VN_HOLD(target_dir_vp);
xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL);
}
if ((src_ip != src_dp) && (src_ip != target_dp)) {
xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL);
}
if ((target_ip != NULL) &&
(target_ip != src_ip) &&
(target_ip != src_dp) &&
(target_ip != target_dp)) {
xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL);
}
/*
* Set up the target.
*/
if (target_ip == NULL) {
/*
* If there's no space reservation, check the entry will
* fit before actually inserting it.
*/
if (spaceres == 0 &&
(error = XFS_DIR_CANENTER(mp, tp, target_dp, target_name,
target_namelen))) {
goto error_return;
}
/*
* If target does not exist and the rename crosses
* directories, adjust the target directory link count
* to account for the ".." reference from the new entry.
*/
error = XFS_DIR_CREATENAME(mp, tp, target_dp, target_name,
target_namelen, src_ip->i_ino,
&first_block, &free_list, spaceres);
if (error == ENOSPC) {
goto error_return;
}
if (error) {
goto abort_return;
}
xfs_ichgtime(target_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
if (new_parent && src_is_directory) {
error = xfs_bumplink(tp, target_dp);
if (error) {
goto abort_return;
}
}
} else { /* target_ip != NULL */
/*
* If target exists and it's a directory, check that both
* target and source are directories and that target can be
* destroyed, or that neither is a directory.
*/
if ((target_ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
/*
* Make sure target dir is empty.
*/
if (!(XFS_DIR_ISEMPTY(target_ip->i_mount, target_ip)) ||
(target_ip->i_d.di_nlink > 2)) {
error = XFS_ERROR(EEXIST);
goto error_return;
}
}
/*
* Link the source inode under the target name.
* If the source inode is a directory and we are moving
* it across directories, its ".." entry will be
* inconsistent until we replace that down below.
*
* In case there is already an entry with the same
* name at the destination directory, remove it first.
*/
error = XFS_DIR_REPLACE(mp, tp, target_dp, target_name,
target_namelen, src_ip->i_ino, &first_block,
&free_list, spaceres);
if (error) {
goto abort_return;
}
xfs_ichgtime(target_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
/*
* Decrement the link count on the target since the target
* dir no longer points to it.
*/
error = xfs_droplink(tp, target_ip);
if (error) {
goto abort_return;
}
target_ip_dropped = 1;
if (src_is_directory) {
/*
* Drop the link from the old "." entry.
*/
error = xfs_droplink(tp, target_ip);
if (error) {
goto abort_return;
}
}
/* Do this test while we still hold the locks */
target_link_zero = (target_ip)->i_d.di_nlink==0;
} /* target_ip != NULL */
/*
* Remove the source.
*/
if (new_parent && src_is_directory) {
/*
* Rewrite the ".." entry to point to the new
* directory.
*/
error = XFS_DIR_REPLACE(mp, tp, src_ip, "..", 2,
target_dp->i_ino, &first_block,
&free_list, spaceres);
ASSERT(error != EEXIST);
if (error) {
goto abort_return;
}
xfs_ichgtime(src_ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
} else {
/*
* We always want to hit the ctime on the source inode.
* We do it in the if clause above for the 'new_parent &&
* src_is_directory' case, and here we get all the other
* cases. This isn't strictly required by the standards
* since the source inode isn't really being changed,
* but old unix file systems did it and some incremental
* backup programs won't work without it.
*/
xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG);
}
/*
* Adjust the link count on src_dp. This is necessary when
* renaming a directory, either within one parent when
* the target existed, or across two parent directories.
*/
if (src_is_directory && (new_parent || target_ip != NULL)) {
/*
* Decrement link count on src_directory since the
* entry that's moved no longer points to it.
*/
error = xfs_droplink(tp, src_dp);
if (error) {
goto abort_return;
}
}
error = XFS_DIR_REMOVENAME(mp, tp, src_dp, src_name, src_namelen,
src_ip->i_ino, &first_block, &free_list, spaceres);
if (error) {
goto abort_return;
}
xfs_ichgtime(src_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
/*
* Update the generation counts on all the directory inodes
* that we're modifying.
*/
src_dp->i_gen++;
xfs_trans_log_inode(tp, src_dp, XFS_ILOG_CORE);
if (new_parent) {
target_dp->i_gen++;
xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE);
}
/*
* If there was a target inode, take an extra reference on
* it here so that it doesn't go to xfs_inactive() from
* within the commit.
*/
if (target_ip != NULL) {
IHOLD(target_ip);
}
/*
* If this is a synchronous mount, make sure that the
* rename transaction goes to disk before returning to
* the user.
*/
if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
xfs_trans_set_sync(tp);
}
/*
* Take refs. for vop_link_removed calls below. No need to worry
* about directory refs. because the caller holds them.
*
* Do holds before the xfs_bmap_finish since it might rele them down
* to zero.
*/
if (target_ip_dropped)
IHOLD(target_ip);
IHOLD(src_ip);
error = xfs_bmap_finish(&tp, &free_list, first_block, &committed);
if (error) {
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES |
XFS_TRANS_ABORT));
if (target_ip != NULL) {
IRELE(target_ip);
}
if (target_ip_dropped) {
IRELE(target_ip);
}
IRELE(src_ip);
goto std_return;
}
/*
* trans_commit will unlock src_ip, target_ip & decrement
* the vnode references.
*/
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES, NULL);
if (target_ip != NULL) {
xfs_refcache_purge_ip(target_ip);
IRELE(target_ip);
}
/*
* Let interposed file systems know about removed links.
*/
if (target_ip_dropped) {
VOP_LINK_REMOVED(XFS_ITOV(target_ip), target_dir_vp,
target_link_zero);
IRELE(target_ip);
}
FSC_NOTIFY_NAME_CHANGED(XFS_ITOV(src_ip));
IRELE(src_ip);
/* Fall through to std_return with error = 0 or errno from
* xfs_trans_commit */
std_return:
if (DM_EVENT_ENABLED(src_dir_vp->v_vfsp, src_dp, DM_EVENT_POSTRENAME) ||
DM_EVENT_ENABLED(target_dir_vp->v_vfsp,
target_dp, DM_EVENT_POSTRENAME)) {
(void) XFS_SEND_NAMESP (mp, DM_EVENT_POSTRENAME,
src_dir_vp, DM_RIGHT_NULL,
target_dir_vp, DM_RIGHT_NULL,
src_name, target_name,
0, error, 0);
}
return error;
abort_return:
cancel_flags |= XFS_TRANS_ABORT;
/* FALLTHROUGH */
error_return:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, cancel_flags);
goto std_return;
rele_return:
IRELE(src_ip);
if (target_ip != NULL) {
IRELE(target_ip);
}
goto std_return;
}
int
xfs_attr_set(
struct xfs_inode *dp,
const unsigned char *name,
unsigned char *value,
int valuelen,
int flags)
{
struct xfs_mount *mp = dp->i_mount;
struct xfs_da_args args;
struct xfs_bmap_free flist;
struct xfs_trans_res tres;
xfs_fsblock_t firstblock;
int rsvd = (flags & ATTR_ROOT) != 0;
int error, err2, committed, local;
XFS_STATS_INC(xs_attr_set);
if (XFS_FORCED_SHUTDOWN(dp->i_mount))
return -EIO;
error = xfs_attr_args_init(&args, dp, name, flags);
if (error)
return error;
args.value = value;
args.valuelen = valuelen;
args.firstblock = &firstblock;
args.flist = &flist;
args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
args.total = xfs_attr_calc_size(&args, &local);
error = xfs_qm_dqattach(dp, 0);
if (error)
return error;
/*
* If the inode doesn't have an attribute fork, add one.
* (inode must not be locked when we call this routine)
*/
if (XFS_IFORK_Q(dp) == 0) {
int sf_size = sizeof(xfs_attr_sf_hdr_t) +
XFS_ATTR_SF_ENTSIZE_BYNAME(args.namelen, valuelen);
error = xfs_bmap_add_attrfork(dp, sf_size, rsvd);
if (error)
return error;
}
/*
* Start our first transaction of the day.
*
* All future transactions during this code must be "chained" off
* this one via the trans_dup() call. All transactions will contain
* the inode, and the inode will always be marked with trans_ihold().
* Since the inode will be locked in all transactions, we must log
* the inode in every transaction to let it float upward through
* the log.
*/
args.trans = xfs_trans_alloc(mp, XFS_TRANS_ATTR_SET);
/*
* Root fork attributes can use reserved data blocks for this
* operation if necessary
*/
if (rsvd)
args.trans->t_flags |= XFS_TRANS_RESERVE;
tres.tr_logres = M_RES(mp)->tr_attrsetm.tr_logres +
M_RES(mp)->tr_attrsetrt.tr_logres * args.total;
tres.tr_logcount = XFS_ATTRSET_LOG_COUNT;
tres.tr_logflags = XFS_TRANS_PERM_LOG_RES;
error = xfs_trans_reserve(args.trans, &tres, args.total, 0);
if (error) {
xfs_trans_cancel(args.trans, 0);
return error;
}
xfs_ilock(dp, XFS_ILOCK_EXCL);
error = xfs_trans_reserve_quota_nblks(args.trans, dp, args.total, 0,
rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
XFS_QMOPT_RES_REGBLKS);
if (error) {
xfs_iunlock(dp, XFS_ILOCK_EXCL);
xfs_trans_cancel(args.trans, XFS_TRANS_RELEASE_LOG_RES);
return error;
}
xfs_trans_ijoin(args.trans, dp, 0);
/*
* If the attribute list is non-existent or a shortform list,
* upgrade it to a single-leaf-block attribute list.
*/
if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL ||
(dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
dp->i_d.di_anextents == 0)) {
/*
* Build initial attribute list (if required).
*/
if (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS)
xfs_attr_shortform_create(&args);
/*
* Try to add the attr to the attribute list in
* the inode.
*/
error = xfs_attr_shortform_addname(&args);
if (error != -ENOSPC) {
/*
* Commit the shortform mods, and we're done.
* NOTE: this is also the error path (EEXIST, etc).
*/
ASSERT(args.trans != NULL);
/*
* If this is a synchronous mount, make sure that
* the transaction goes to disk before returning
* to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(args.trans);
if (!error && (flags & ATTR_KERNOTIME) == 0) {
xfs_trans_ichgtime(args.trans, dp,
XFS_ICHGTIME_CHG);
}
err2 = xfs_trans_commit(args.trans,
XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error ? error : err2;
}
/*
* It won't fit in the shortform, transform to a leaf block.
* GROT: another possible req'mt for a double-split btree op.
*/
xfs_bmap_init(args.flist, args.firstblock);
error = xfs_attr_shortform_to_leaf(&args);
if (!error) {
error = xfs_bmap_finish(&args.trans, args.flist,
&committed);
}
if (error) {
ASSERT(committed);
args.trans = NULL;
xfs_bmap_cancel(&flist);
goto out;
}
/*
* bmap_finish() may have committed the last trans and started
* a new one. We need the inode to be in all transactions.
*/
if (committed)
xfs_trans_ijoin(args.trans, dp, 0);
/*
* Commit the leaf transformation. We'll need another (linked)
* transaction to add the new attribute to the leaf.
*/
error = xfs_trans_roll(&args.trans, dp);
if (error)
goto out;
}
if (xfs_bmap_one_block(dp, XFS_ATTR_FORK))
error = xfs_attr_leaf_addname(&args);
else
error = xfs_attr_node_addname(&args);
if (error)
goto out;
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(args.trans);
if ((flags & ATTR_KERNOTIME) == 0)
xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
/*
* Commit the last in the sequence of transactions.
*/
xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
out:
if (args.trans) {
xfs_trans_cancel(args.trans,
XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
}
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
}
/*
* Allocates a new inode from disk and return a pointer to the
* incore copy. This routine will internally commit the current
* transaction and allocate a new one if the Space Manager needed
* to do an allocation to replenish the inode free-list.
*
* This routine is designed to be called from xfs_create and
* xfs_create_dir.
*
*/
int
xfs_dir_ialloc(
xfs_trans_t **tpp, /* input: current transaction;
output: may be a new transaction. */
xfs_inode_t *dp, /* directory within whose allocate
the inode. */
umode_t mode,
xfs_nlink_t nlink,
xfs_dev_t rdev,
prid_t prid, /* project id */
int okalloc, /* ok to allocate new space */
xfs_inode_t **ipp, /* pointer to inode; it will be
locked. */
int *committed)
{
xfs_trans_t *tp;
xfs_trans_t *ntp;
xfs_inode_t *ip;
xfs_buf_t *ialloc_context = NULL;
int code;
uint log_res;
uint log_count;
void *dqinfo;
uint tflags;
tp = *tpp;
ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
/*
* xfs_ialloc will return a pointer to an incore inode if
* the Space Manager has an available inode on the free
* list. Otherwise, it will do an allocation and replenish
* the freelist. Since we can only do one allocation per
* transaction without deadlocks, we will need to commit the
* current transaction and start a new one. We will then
* need to call xfs_ialloc again to get the inode.
*
* If xfs_ialloc did an allocation to replenish the freelist,
* it returns the bp containing the head of the freelist as
* ialloc_context. We will hold a lock on it across the
* transaction commit so that no other process can steal
* the inode(s) that we've just allocated.
*/
code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid, okalloc,
&ialloc_context, &ip);
/*
* Return an error if we were unable to allocate a new inode.
* This should only happen if we run out of space on disk or
* encounter a disk error.
*/
if (code) {
*ipp = NULL;
return code;
}
if (!ialloc_context && !ip) {
*ipp = NULL;
return XFS_ERROR(ENOSPC);
}
/*
* If the AGI buffer is non-NULL, then we were unable to get an
* inode in one operation. We need to commit the current
* transaction and call xfs_ialloc() again. It is guaranteed
* to succeed the second time.
*/
if (ialloc_context) {
/*
* Normally, xfs_trans_commit releases all the locks.
* We call bhold to hang on to the ialloc_context across
* the commit. Holding this buffer prevents any other
* processes from doing any allocations in this
* allocation group.
*/
xfs_trans_bhold(tp, ialloc_context);
/*
* Save the log reservation so we can use
* them in the next transaction.
*/
log_res = xfs_trans_get_log_res(tp);
log_count = xfs_trans_get_log_count(tp);
/*
* We want the quota changes to be associated with the next
* transaction, NOT this one. So, detach the dqinfo from this
* and attach it to the next transaction.
*/
dqinfo = NULL;
tflags = 0;
if (tp->t_dqinfo) {
dqinfo = (void *)tp->t_dqinfo;
tp->t_dqinfo = NULL;
tflags = tp->t_flags & XFS_TRANS_DQ_DIRTY;
tp->t_flags &= ~(XFS_TRANS_DQ_DIRTY);
}
ntp = xfs_trans_dup(tp);
code = xfs_trans_commit(tp, 0);
tp = ntp;
if (committed != NULL) {
*committed = 1;
}
/*
* If we get an error during the commit processing,
* release the buffer that is still held and return
* to the caller.
*/
if (code) {
xfs_buf_relse(ialloc_context);
if (dqinfo) {
tp->t_dqinfo = dqinfo;
xfs_trans_free_dqinfo(tp);
}
*tpp = ntp;
*ipp = NULL;
return code;
}
/*
* transaction commit worked ok so we can drop the extra ticket
* reference that we gained in xfs_trans_dup()
*/
xfs_log_ticket_put(tp->t_ticket);
code = xfs_trans_reserve(tp, 0, log_res, 0,
XFS_TRANS_PERM_LOG_RES, log_count);
/*
* Re-attach the quota info that we detached from prev trx.
*/
if (dqinfo) {
tp->t_dqinfo = dqinfo;
tp->t_flags |= tflags;
}
if (code) {
xfs_buf_relse(ialloc_context);
*tpp = ntp;
*ipp = NULL;
return code;
}
xfs_trans_bjoin(tp, ialloc_context);
/*
* Call ialloc again. Since we've locked out all
* other allocations in this allocation group,
* this call should always succeed.
*/
code = xfs_ialloc(tp, dp, mode, nlink, rdev, prid,
okalloc, &ialloc_context, &ip);
/*
* If we get an error at this point, return to the caller
* so that the current transaction can be aborted.
*/
if (code) {
*tpp = tp;
*ipp = NULL;
return code;
}
ASSERT(!ialloc_context && ip);
} else {
if (committed != NULL)
*committed = 0;
}
*ipp = ip;
*tpp = tp;
return 0;
}
/*
* Try to truncate the given file to 0 length. Currently called
* only out of xfs_remove when it has to truncate a file to free
* up space for the remove to proceed.
*/
int
xfs_truncate_file(
xfs_mount_t *mp,
xfs_inode_t *ip)
{
xfs_trans_t *tp;
int error;
#ifdef QUOTADEBUG
/*
* This is called to truncate the quotainodes too.
*/
if (XFS_IS_UQUOTA_ON(mp)) {
if (ip->i_ino != mp->m_sb.sb_uquotino)
ASSERT(ip->i_udquot);
}
if (XFS_IS_OQUOTA_ON(mp)) {
if (ip->i_ino != mp->m_sb.sb_gquotino)
ASSERT(ip->i_gdquot);
}
#endif
/*
* Make the call to xfs_itruncate_start before starting the
* transaction, because we cannot make the call while we're
* in a transaction.
*/
xfs_ilock(ip, XFS_IOLOCK_EXCL);
xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE, (xfs_fsize_t)0);
tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE);
if ((error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_ITRUNCATE_LOG_COUNT))) {
xfs_trans_cancel(tp, 0);
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return error;
}
/*
* Follow the normal truncate locking protocol. Since we
* hold the inode in the transaction, we know that it's number
* of references will stay constant.
*/
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Signal a sync xaction. The only case where that isn't
* the case is if we're truncating an already unlinked file
* on a wsync fs. In that case, we know the blocks can't
* reappear in the file because the links to file are
* permanently toast. Currently, we're always going to
* want a sync transaction because this code is being
* called from places where nlink is guaranteed to be 1
* but I'm leaving the tests in to protect against future
* changes -- rcc.
*/
error = xfs_itruncate_finish(&tp, ip, (xfs_fsize_t)0,
XFS_DATA_FORK,
((ip->i_d.di_nlink != 0 ||
!(mp->m_flags & XFS_MOUNT_WSYNC))
? 1 : 0));
if (error) {
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
XFS_TRANS_ABORT);
} else {
xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES,
NULL);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
return error;
}
STATIC int
xfs_file_fsync(
struct file *file,
int datasync)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
int error = 0;
int log_flushed = 0;
trace_xfs_file_fsync(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return -XFS_ERROR(EIO);
xfs_iflags_clear(ip, XFS_ITRUNCATED);
xfs_ioend_wait(ip);
if (mp->m_flags & XFS_MOUNT_BARRIER) {
/*
* If we have an RT and/or log subvolume we need to make sure
* to flush the write cache the device used for file data
* first. This is to ensure newly written file data make
* it to disk before logging the new inode size in case of
* an extending write.
*/
if (XFS_IS_REALTIME_INODE(ip))
xfs_blkdev_issue_flush(mp->m_rtdev_targp);
else if (mp->m_logdev_targp != mp->m_ddev_targp)
xfs_blkdev_issue_flush(mp->m_ddev_targp);
}
/*
* We always need to make sure that the required inode state is safe on
* disk. The inode might be clean but we still might need to force the
* log because of committed transactions that haven't hit the disk yet.
* Likewise, there could be unflushed non-transactional changes to the
* inode core that have to go to disk and this requires us to issue
* a synchronous transaction to capture these changes correctly.
*
* This code relies on the assumption that if the i_update_core field
* of the inode is clear and the inode is unpinned then it is clean
* and no action is required.
*/
xfs_ilock(ip, XFS_ILOCK_SHARED);
/*
* First check if the VFS inode is marked dirty. All the dirtying
* of non-transactional updates no goes through mark_inode_dirty*,
* which allows us to distinguish beteeen pure timestamp updates
* and i_size updates which need to be caught for fdatasync.
* After that also theck for the dirty state in the XFS inode, which
* might gets cleared when the inode gets written out via the AIL
* or xfs_iflush_cluster.
*/
if (((inode->i_state & I_DIRTY_DATASYNC) ||
((inode->i_state & I_DIRTY_SYNC) && !datasync)) &&
ip->i_update_core) {
/*
* Kick off a transaction to log the inode core to get the
* updates. The sync transaction will also force the log.
*/
xfs_iunlock(ip, XFS_ILOCK_SHARED);
tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
error = xfs_trans_reserve(tp, 0,
XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
if (error) {
xfs_trans_cancel(tp, 0);
return -error;
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
/*
* Note - it's possible that we might have pushed ourselves out
* of the way during trans_reserve which would flush the inode.
* But there's no guarantee that the inode buffer has actually
* gone out yet (it's delwri). Plus the buffer could be pinned
* anyway if it's part of an inode in another recent
* transaction. So we play it safe and fire off the
* transaction anyway.
*/
xfs_trans_ijoin(tp, ip);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
xfs_trans_set_sync(tp);
error = _xfs_trans_commit(tp, 0, &log_flushed);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
} else {
/*
* Timestamps/size haven't changed since last inode flush or
* inode transaction commit. That means either nothing got
* written or a transaction committed which caught the updates.
* If the latter happened and the transaction hasn't hit the
* disk yet, the inode will be still be pinned. If it is,
* force the log.
*/
if (xfs_ipincount(ip)) {
error = _xfs_log_force_lsn(mp,
ip->i_itemp->ili_last_lsn,
XFS_LOG_SYNC, &log_flushed);
}
xfs_iunlock(ip, XFS_ILOCK_SHARED);
}
/*
* If we only have a single device, and the log force about was
* a no-op we might have to flush the data device cache here.
* This can only happen for fdatasync/O_DSYNC if we were overwriting
* an already allocated file and thus do not have any metadata to
* commit.
*/
if ((mp->m_flags & XFS_MOUNT_BARRIER) &&
mp->m_logdev_targp == mp->m_ddev_targp &&
!XFS_IS_REALTIME_INODE(ip) &&
!log_flushed)
xfs_blkdev_issue_flush(mp->m_ddev_targp);
return -error;
}
int
xfs_setattr_nonsize(
struct xfs_inode *ip,
struct iattr *iattr,
int flags)
{
xfs_mount_t *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
xfs_trans_t *tp;
int error;
uid_t uid = 0, iuid = 0;
gid_t gid = 0, igid = 0;
struct xfs_dquot *udqp = NULL, *gdqp = NULL;
struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
ASSERT((mask & ATTR_SIZE) == 0);
/*
* If disk quotas is on, we make sure that the dquots do exist on disk,
* before we start any other transactions. Trying to do this later
* is messy. We don't care to take a readlock to look at the ids
* in inode here, because we can't hold it across the trans_reserve.
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) {
uint qflags = 0;
if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) {
uid = iattr->ia_uid;
qflags |= XFS_QMOPT_UQUOTA;
} else {
uid = ip->i_d.di_uid;
}
if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) {
gid = iattr->ia_gid;
qflags |= XFS_QMOPT_GQUOTA;
} else {
gid = ip->i_d.di_gid;
}
/*
* We take a reference when we initialize udqp and gdqp,
* so it is important that we never blindly double trip on
* the same variable. See xfs_create() for an example.
*/
ASSERT(udqp == NULL);
ASSERT(gdqp == NULL);
error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip),
qflags, &udqp, &gdqp);
if (error)
return error;
}
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
if (error)
goto out_dqrele;
xfs_ilock(ip, XFS_ILOCK_EXCL);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & (ATTR_UID|ATTR_GID)) {
/*
* These IDs could have changed since we last looked at them.
* But, we're assured that if the ownership did change
* while we didn't have the inode locked, inode's dquot(s)
* would have changed also.
*/
iuid = ip->i_d.di_uid;
igid = ip->i_d.di_gid;
gid = (mask & ATTR_GID) ? iattr->ia_gid : igid;
uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid;
/*
* Do a quota reservation only if uid/gid is actually
* going to change.
*/
if (XFS_IS_QUOTA_RUNNING(mp) &&
((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
(XFS_IS_GQUOTA_ON(mp) && igid != gid))) {
ASSERT(tp);
error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (error) /* out of quota */
goto out_trans_cancel;
}
}
xfs_trans_ijoin(tp, ip);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & (ATTR_UID|ATTR_GID)) {
/*
* CAP_FSETID overrides the following restrictions:
*
* The set-user-ID and set-group-ID bits of a file will be
* cleared upon successful return from chown()
*/
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
/*
* Change the ownerships and register quota modifications
* in the transaction.
*/
if (iuid != uid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) {
ASSERT(mask & ATTR_UID);
ASSERT(udqp);
olddquot1 = xfs_qm_vop_chown(tp, ip,
&ip->i_udquot, udqp);
}
ip->i_d.di_uid = uid;
inode->i_uid = uid;
}
if (igid != gid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
ASSERT(!XFS_IS_PQUOTA_ON(mp));
ASSERT(mask & ATTR_GID);
ASSERT(gdqp);
olddquot2 = xfs_qm_vop_chown(tp, ip,
&ip->i_gdquot, gdqp);
}
ip->i_d.di_gid = gid;
inode->i_gid = gid;
}
}
/*
* Change file access modes.
*/
if (mask & ATTR_MODE) {
umode_t mode = iattr->ia_mode;
if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
mode &= ~S_ISGID;
ip->i_d.di_mode &= S_IFMT;
ip->i_d.di_mode |= mode & ~S_IFMT;
inode->i_mode &= S_IFMT;
inode->i_mode |= mode & ~S_IFMT;
}
/*
* Change file access or modified times.
*/
if (mask & ATTR_ATIME) {
inode->i_atime = iattr->ia_atime;
ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
ip->i_update_core = 1;
}
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
ip->i_update_core = 1;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
ip->i_update_core = 1;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/*
* Release any dquot(s) the inode had kept before chown.
*/
xfs_qm_dqrele(olddquot1);
xfs_qm_dqrele(olddquot2);
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
if (error)
return XFS_ERROR(error);
/*
* XXX(hch): Updating the ACL entries is not atomic vs the i_mode
* update. We could avoid this with linked transactions
* and passing down the transaction pointer all the way
* to attr_set. No previous user of the generic
* Posix ACL code seems to care about this issue either.
*/
if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
error = -xfs_acl_chmod(inode);
if (error)
return XFS_ERROR(error);
}
return 0;
out_trans_cancel:
xfs_trans_cancel(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
out_dqrele:
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
return error;
}
/*
* Syssgi interface for swapext
*/
int
xfs_swapext(
xfs_swapext_t __user *sxp)
{
xfs_swapext_t sx;
xfs_inode_t *ip=NULL, *tip=NULL, *ips[2];
xfs_trans_t *tp;
xfs_mount_t *mp;
xfs_bstat_t *sbp;
struct file *fp = NULL, *tfp = NULL;
vnode_t *vp, *tvp;
bhv_desc_t *bdp, *tbdp;
vn_bhv_head_t *bhp, *tbhp;
uint lock_flags=0;
int ilf_fields, tilf_fields;
int error = 0;
xfs_ifork_t tempif, *ifp, *tifp;
__uint64_t tmp;
int aforkblks = 0;
int taforkblks = 0;
int locked = 0;
if (copy_from_user(&sx, sxp, sizeof(sx)))
return XFS_ERROR(EFAULT);
/* Pull information for the target fd */
if (((fp = fget((int)sx.sx_fdtarget)) == NULL) ||
((vp = LINVFS_GET_VP(fp->f_dentry->d_inode)) == NULL)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
bhp = VN_BHV_HEAD(vp);
bdp = vn_bhv_lookup(bhp, &xfs_vnodeops);
if (bdp == NULL) {
error = XFS_ERROR(EBADF);
goto error0;
} else {
ip = XFS_BHVTOI(bdp);
}
if (((tfp = fget((int)sx.sx_fdtmp)) == NULL) ||
((tvp = LINVFS_GET_VP(tfp->f_dentry->d_inode)) == NULL)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
tbhp = VN_BHV_HEAD(tvp);
tbdp = vn_bhv_lookup(tbhp, &xfs_vnodeops);
if (tbdp == NULL) {
error = XFS_ERROR(EBADF);
goto error0;
} else {
tip = XFS_BHVTOI(tbdp);
}
if (ip->i_mount != tip->i_mount) {
error = XFS_ERROR(EINVAL);
goto error0;
}
if (ip->i_ino == tip->i_ino) {
error = XFS_ERROR(EINVAL);
goto error0;
}
mp = ip->i_mount;
sbp = &sx.sx_stat;
if (XFS_FORCED_SHUTDOWN(mp)) {
error = XFS_ERROR(EIO);
goto error0;
}
locked = 1;
/* Lock in i_ino order */
if (ip->i_ino < tip->i_ino) {
ips[0] = ip;
ips[1] = tip;
} else {
ips[0] = tip;
ips[1] = ip;
}
lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL;
xfs_lock_inodes(ips, 2, 0, lock_flags);
/* Check permissions */
error = xfs_iaccess(ip, S_IWUSR, NULL);
if (error)
goto error0;
error = xfs_iaccess(tip, S_IWUSR, NULL);
if (error)
goto error0;
/* Verify that both files have the same format */
if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
/* Verify both files are either real-time or non-realtime */
if ((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) !=
(tip->i_d.di_flags & XFS_DIFLAG_REALTIME)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
/* Should never get a local format */
if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
error = XFS_ERROR(EINVAL);
goto error0;
}
if (VN_CACHED(tvp) != 0)
xfs_inval_cached_pages(XFS_ITOV(tip), &(tip->i_iocore),
(xfs_off_t)0, 0, 0);
/* Verify O_DIRECT for ftmp */
if (VN_CACHED(tvp) != 0) {
error = XFS_ERROR(EINVAL);
goto error0;
}
/* Verify all data are being swapped */
if (sx.sx_offset != 0 ||
sx.sx_length != ip->i_d.di_size ||
sx.sx_length != tip->i_d.di_size) {
error = XFS_ERROR(EFAULT);
goto error0;
}
/*
* If the target has extended attributes, the tmp file
* must also in order to ensure the correct data fork
* format.
*/
if ( XFS_IFORK_Q(ip) != XFS_IFORK_Q(tip) ) {
error = XFS_ERROR(EINVAL);
goto error0;
}
/*
* Compare the current change & modify times with that
* passed in. If they differ, we abort this swap.
* This is the mechanism used to ensure the calling
* process that the file was not changed out from
* under it.
*/
if ((sbp->bs_ctime.tv_sec != ip->i_d.di_ctime.t_sec) ||
(sbp->bs_ctime.tv_nsec != ip->i_d.di_ctime.t_nsec) ||
(sbp->bs_mtime.tv_sec != ip->i_d.di_mtime.t_sec) ||
(sbp->bs_mtime.tv_nsec != ip->i_d.di_mtime.t_nsec)) {
error = XFS_ERROR(EBUSY);
goto error0;
}
/* We need to fail if the file is memory mapped. Once we have tossed
* all existing pages, the page fault will have no option
* but to go to the filesystem for pages. By making the page fault call
* VOP_READ (or write in the case of autogrow) they block on the iolock
* until we have switched the extents.
*/
if (VN_MAPPED(vp)) {
error = XFS_ERROR(EBUSY);
goto error0;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_iunlock(tip, XFS_ILOCK_EXCL);
/*
* There is a race condition here since we gave up the
* ilock. However, the data fork will not change since
* we have the iolock (locked for truncation too) so we
* are safe. We don't really care if non-io related
* fields change.
*/
VOP_TOSS_PAGES(vp, 0, -1, FI_REMAPF);
tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
if ((error = xfs_trans_reserve(tp, 0,
XFS_ICHANGE_LOG_RES(mp), 0,
0, 0))) {
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
xfs_iunlock(tip, XFS_IOLOCK_EXCL);
xfs_trans_cancel(tp, 0);
return error;
}
xfs_lock_inodes(ips, 2, 0, XFS_ILOCK_EXCL);
/*
* Count the number of extended attribute blocks
*/
if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) &&
(ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks);
if (error) {
xfs_iunlock(ip, lock_flags);
xfs_iunlock(tip, lock_flags);
xfs_trans_cancel(tp, 0);
return error;
}
}
if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) &&
(tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK,
&taforkblks);
if (error) {
xfs_iunlock(ip, lock_flags);
xfs_iunlock(tip, lock_flags);
xfs_trans_cancel(tp, 0);
return error;
}
}
/*
* Swap the data forks of the inodes
*/
ifp = &ip->i_df;
tifp = &tip->i_df;
tempif = *ifp; /* struct copy */
*ifp = *tifp; /* struct copy */
*tifp = tempif; /* struct copy */
/*
* Fix the on-disk inode values
*/
tmp = (__uint64_t)ip->i_d.di_nblocks;
ip->i_d.di_nblocks = tip->i_d.di_nblocks - taforkblks + aforkblks;
tip->i_d.di_nblocks = tmp + taforkblks - aforkblks;
tmp = (__uint64_t) ip->i_d.di_nextents;
ip->i_d.di_nextents = tip->i_d.di_nextents;
tip->i_d.di_nextents = tmp;
tmp = (__uint64_t) ip->i_d.di_format;
ip->i_d.di_format = tip->i_d.di_format;
tip->i_d.di_format = tmp;
ilf_fields = XFS_ILOG_CORE;
switch(ip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (ip->i_d.di_nextents <= XFS_INLINE_EXTS) {
ifp->if_u1.if_extents =
ifp->if_u2.if_inline_ext;
}
ilf_fields |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
ilf_fields |= XFS_ILOG_DBROOT;
break;
}
tilf_fields = XFS_ILOG_CORE;
switch(tip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (tip->i_d.di_nextents <= XFS_INLINE_EXTS) {
tifp->if_u1.if_extents =
tifp->if_u2.if_inline_ext;
}
tilf_fields |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
tilf_fields |= XFS_ILOG_DBROOT;
break;
}
/*
* Increment vnode ref counts since xfs_trans_commit &
* xfs_trans_cancel will both unlock the inodes and
* decrement the associated ref counts.
*/
VN_HOLD(vp);
VN_HOLD(tvp);
xfs_trans_ijoin(tp, ip, lock_flags);
xfs_trans_ijoin(tp, tip, lock_flags);
xfs_trans_log_inode(tp, ip, ilf_fields);
xfs_trans_log_inode(tp, tip, tilf_fields);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC) {
xfs_trans_set_sync(tp);
}
error = xfs_trans_commit(tp, XFS_TRANS_SWAPEXT, NULL);
fput(fp);
fput(tfp);
return error;
error0:
if (locked) {
xfs_iunlock(ip, lock_flags);
xfs_iunlock(tip, lock_flags);
}
if (fp != NULL) fput(fp);
if (tfp != NULL) fput(tfp);
return error;
}
int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int flags,
xfs_bmbt_irec_t *ret_imap,
int *nmaps,
int found)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_filblks_t count_fsb, resaligned;
xfs_fsblock_t firstfsb;
xfs_extlen_t extsz, temp;
int nimaps;
int bmapi_flag;
int quota_flag;
int rt;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint qblocks, resblks, resrtextents;
int committed;
int error;
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = xfs_qm_dqattach_locked(ip, 0);
if (error)
return XFS_ERROR(error);
rt = XFS_IS_REALTIME_INODE(ip);
extsz = xfs_get_extsz_hint(ip);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
if ((offset + count) > ip->i_size) {
error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
if (error)
goto error_out;
} else {
if (found && (ret_imap->br_startblock == HOLESTARTBLOCK))
last_fsb = MIN(last_fsb, (xfs_fileoff_t)
ret_imap->br_blockcount +
ret_imap->br_startoff);
}
count_fsb = last_fsb - offset_fsb;
ASSERT(count_fsb > 0);
resaligned = count_fsb;
if (unlikely(extsz)) {
if ((temp = do_mod(offset_fsb, extsz)))
resaligned += temp;
if ((temp = do_mod(resaligned, extsz)))
resaligned += extsz - temp;
}
if (unlikely(rt)) {
resrtextents = qblocks = resaligned;
resrtextents /= mp->m_sb.sb_rextsize;
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
quota_flag = XFS_QMOPT_RES_RTBLKS;
} else {
resrtextents = 0;
resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
quota_flag = XFS_QMOPT_RES_REGBLKS;
}
/*
* Allocate and setup the transaction
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), resrtextents,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
/*
* Check for running out of space, note: need lock to return
*/
if (error)
xfs_trans_cancel(tp, 0);
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
goto error_out;
error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
if (error)
goto error1;
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
bmapi_flag = XFS_BMAPI_WRITE;
if ((flags & BMAPI_DIRECT) && (offset < ip->i_size || extsz))
bmapi_flag |= XFS_BMAPI_PREALLOC;
/*
* Issue the xfs_bmapi() call to allocate the blocks
*/
xfs_bmap_init(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, bmapi_flag,
&firstfsb, 0, &imap, &nimaps, &free_list, NULL);
if (error)
goto error0;
/*
* Complete the transaction
*/
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto error0;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error_out;
/*
* Copy any maps to caller's array and return any error.
*/
if (nimaps == 0) {
error = ENOSPC;
goto error_out;
}
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip))) {
error = xfs_cmn_err_fsblock_zero(ip, &imap);
goto error_out;
}
*ret_imap = imap;
*nmaps = 1;
return 0;
error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
xfs_bmap_cancel(&free_list);
xfs_trans_unreserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
error1: /* Just cancel transaction */
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
*nmaps = 0; /* nothing set-up here */
error_out:
return XFS_ERROR(error);
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*
* We no longer bother to look at the incoming map - all we have to
* guarantee is that whatever we allocate fills the required range.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap;
xfs_trans_t *tp;
int nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
error = xfs_qm_dqattach(ip, 0);
if (error)
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
xfs_bmap_init(&free_list, &first_block);
/*
* it is possible that the extents have changed since
* we did the read call as we dropped the ilock for a
* while. We have to be careful about truncates or hole
* punchs here - we are not allowed to allocate
* non-delalloc blocks here.
*
* The only protection against truncation is the pages
* for the range we are being asked to convert are
* locked and hence a truncate will block on them
* first.
*
* As a result, if we go beyond the range we really
* need and hit an delalloc extent boundary followed by
* a hole while we have excess blocks in the map, we
* will fill the hole incorrectly and overrun the
* transaction reservation.
*
* Using a single map prevents this as we are forced to
* check each map we look for overlap with the desired
* range and abort as soon as we find it. Also, given
* that we only return a single map, having one beyond
* what we can return is probably a bit silly.
*
* We also need to check that we don't go beyond EOF;
* this is a truncate optimisation as a truncate sets
* the new file size before block on the pages we
* currently have locked under writeback. Because they
* are about to be tossed, we don't need to write them
* back....
*/
nimaps = 1;
end_fsb = XFS_B_TO_FSB(mp, ip->i_size);
error = xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
if (error)
goto trans_cancel;
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
&imap, &nimaps, &free_list, NULL);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((offset_fsb >= imap.br_startoff) &&
(offset_fsb < (imap.br_startoff +
imap.br_blockcount))) {
*map = imap;
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
/*
* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
count_fsb -= imap.br_blockcount;
map_start_fsb = imap.br_startoff + imap.br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* allocate an incore dquot from the kernel heap,
* and fill its core with quota information kept on disk.
* If XFS_QMOPT_DQALLOC is set, it'll allocate a dquot on disk
* if it wasn't already allocated.
*/
STATIC int
xfs_qm_idtodq(
xfs_mount_t *mp,
xfs_dqid_t id, /* gid or uid, depending on type */
uint type, /* UDQUOT or GDQUOT */
uint flags, /* DQALLOC, DQREPAIR */
xfs_dquot_t **O_dqpp)/* OUT : incore dquot, not locked */
{
xfs_dquot_t *dqp;
int error;
xfs_trans_t *tp;
int cancelflags=0;
dqp = xfs_qm_dqinit(mp, id, type);
tp = NULL;
if (flags & XFS_QMOPT_DQALLOC) {
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
if ((error = xfs_trans_reserve(tp,
XFS_QM_DQALLOC_SPACE_RES(mp),
XFS_WRITE_LOG_RES(mp) +
BBTOB(XFS_QI_DQCHUNKLEN(mp)) - 1 +
128,
0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT))) {
cancelflags = 0;
goto error0;
}
cancelflags = XFS_TRANS_RELEASE_LOG_RES;
}
/*
* Read it from disk; xfs_dqread() takes care of
* all the necessary initialization of dquot's fields (locks, etc)
*/
if ((error = xfs_qm_dqread(tp, id, dqp, flags))) {
/*
* This can happen if quotas got turned off (ESRCH),
* or if the dquot didn't exist on disk and we ask to
* allocate (ENOENT).
*/
xfs_dqtrace_entry(dqp, "DQREAD FAIL");
cancelflags |= XFS_TRANS_ABORT;
goto error0;
}
if (tp) {
if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES,
NULL)))
goto error1;
}
*O_dqpp = dqp;
return (0);
error0:
ASSERT(error);
if (tp)
xfs_trans_cancel(tp, cancelflags);
error1:
xfs_qm_dqdestroy(dqp);
*O_dqpp = NULL;
return (error);
}
/*
* Adjust quota limits, and start/stop timers accordingly.
*/
int
xfs_qm_scall_setqlim(
xfs_mount_t *mp,
xfs_dqid_t id,
uint type,
fs_disk_quota_t *newlim)
{
struct xfs_quotainfo *q = mp->m_quotainfo;
xfs_disk_dquot_t *ddq;
xfs_dquot_t *dqp;
xfs_trans_t *tp;
int error;
xfs_qcnt_t hard, soft;
if (newlim->d_fieldmask & ~XFS_DQ_MASK)
return EINVAL;
if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
return 0;
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128,
0, 0, XFS_DEFAULT_LOG_COUNT))) {
xfs_trans_cancel(tp, 0);
return (error);
}
/*
* We don't want to race with a quotaoff so take the quotaoff lock.
* (We don't hold an inode lock, so there's nothing else to stop
* a quotaoff from happening). (XXXThis doesn't currently happen
* because we take the vfslock before calling xfs_qm_sysent).
*/
mutex_lock(&q->qi_quotaofflock);
/*
* Get the dquot (locked), and join it to the transaction.
* Allocate the dquot if this doesn't exist.
*/
if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) {
xfs_trans_cancel(tp, XFS_TRANS_ABORT);
ASSERT(error != ENOENT);
goto out_unlock;
}
xfs_trans_dqjoin(tp, dqp);
ddq = &dqp->q_core;
/*
* Make sure that hardlimits are >= soft limits before changing.
*/
hard = (newlim->d_fieldmask & FS_DQ_BHARD) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_hardlimit) :
be64_to_cpu(ddq->d_blk_hardlimit);
soft = (newlim->d_fieldmask & FS_DQ_BSOFT) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_softlimit) :
be64_to_cpu(ddq->d_blk_softlimit);
if (hard == 0 || hard >= soft) {
ddq->d_blk_hardlimit = cpu_to_be64(hard);
ddq->d_blk_softlimit = cpu_to_be64(soft);
if (id == 0) {
q->qi_bhardlimit = hard;
q->qi_bsoftlimit = soft;
}
} else {
xfs_debug(mp, "blkhard %Ld < blksoft %Ld\n", hard, soft);
}
hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) :
be64_to_cpu(ddq->d_rtb_hardlimit);
soft = (newlim->d_fieldmask & FS_DQ_RTBSOFT) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_softlimit) :
be64_to_cpu(ddq->d_rtb_softlimit);
if (hard == 0 || hard >= soft) {
ddq->d_rtb_hardlimit = cpu_to_be64(hard);
ddq->d_rtb_softlimit = cpu_to_be64(soft);
if (id == 0) {
q->qi_rtbhardlimit = hard;
q->qi_rtbsoftlimit = soft;
}
} else {
xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld\n", hard, soft);
}
hard = (newlim->d_fieldmask & FS_DQ_IHARD) ?
(xfs_qcnt_t) newlim->d_ino_hardlimit :
be64_to_cpu(ddq->d_ino_hardlimit);
soft = (newlim->d_fieldmask & FS_DQ_ISOFT) ?
(xfs_qcnt_t) newlim->d_ino_softlimit :
be64_to_cpu(ddq->d_ino_softlimit);
if (hard == 0 || hard >= soft) {
ddq->d_ino_hardlimit = cpu_to_be64(hard);
ddq->d_ino_softlimit = cpu_to_be64(soft);
if (id == 0) {
q->qi_ihardlimit = hard;
q->qi_isoftlimit = soft;
}
} else {
xfs_debug(mp, "ihard %Ld < isoft %Ld\n", hard, soft);
}
/*
* Update warnings counter(s) if requested
*/
if (newlim->d_fieldmask & FS_DQ_BWARNS)
ddq->d_bwarns = cpu_to_be16(newlim->d_bwarns);
if (newlim->d_fieldmask & FS_DQ_IWARNS)
ddq->d_iwarns = cpu_to_be16(newlim->d_iwarns);
if (newlim->d_fieldmask & FS_DQ_RTBWARNS)
ddq->d_rtbwarns = cpu_to_be16(newlim->d_rtbwarns);
if (id == 0) {
/*
* Timelimits for the super user set the relative time
* the other users can be over quota for this file system.
* If it is zero a default is used. Ditto for the default
* soft and hard limit values (already done, above), and
* for warnings.
*/
if (newlim->d_fieldmask & FS_DQ_BTIMER) {
q->qi_btimelimit = newlim->d_btimer;
ddq->d_btimer = cpu_to_be32(newlim->d_btimer);
}
if (newlim->d_fieldmask & FS_DQ_ITIMER) {
q->qi_itimelimit = newlim->d_itimer;
ddq->d_itimer = cpu_to_be32(newlim->d_itimer);
}
if (newlim->d_fieldmask & FS_DQ_RTBTIMER) {
q->qi_rtbtimelimit = newlim->d_rtbtimer;
ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer);
}
if (newlim->d_fieldmask & FS_DQ_BWARNS)
q->qi_bwarnlimit = newlim->d_bwarns;
if (newlim->d_fieldmask & FS_DQ_IWARNS)
q->qi_iwarnlimit = newlim->d_iwarns;
if (newlim->d_fieldmask & FS_DQ_RTBWARNS)
q->qi_rtbwarnlimit = newlim->d_rtbwarns;
} else {
/*
* If the user is now over quota, start the timelimit.
* The user will not be 'warned'.
* Note that we keep the timers ticking, whether enforcement
* is on or off. We don't really want to bother with iterating
* over all ondisk dquots and turning the timers on/off.
*/
xfs_qm_adjust_dqtimers(mp, ddq);
}
dqp->dq_flags |= XFS_DQ_DIRTY;
xfs_trans_log_dquot(tp, dqp);
error = xfs_trans_commit(tp, 0);
xfs_qm_dqrele(dqp);
out_unlock:
mutex_unlock(&q->qi_quotaofflock);
return error;
}
/*
* Truncate file. Must have write permission and not be a directory.
*/
int
xfs_setattr_size(
struct xfs_inode *ip,
struct iattr *iattr)
{
struct xfs_mount *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
xfs_off_t oldsize, newsize;
struct xfs_trans *tp;
int error;
uint lock_flags = 0;
bool did_zeroing = false;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return -EROFS;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
error = inode_change_ok(inode, iattr);
if (error)
return error;
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
ASSERT(xfs_isilocked(ip, XFS_MMAPLOCK_EXCL));
ASSERT(S_ISREG(ip->i_d.di_mode));
ASSERT((iattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
oldsize = inode->i_size;
newsize = iattr->ia_size;
/*
* Short circuit the truncate case for zero length files.
*/
if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
if (!(iattr->ia_valid & (ATTR_CTIME|ATTR_MTIME)))
return 0;
/*
* Use the regular setattr path to update the timestamps.
*/
iattr->ia_valid &= ~ATTR_SIZE;
return xfs_setattr_nonsize(ip, iattr, 0);
}
/*
* Make sure that the dquots are attached to the inode.
*/
error = xfs_qm_dqattach(ip, 0);
if (error)
return error;
/*
* File data changes must be complete before we start the transaction to
* modify the inode. This needs to be done before joining the inode to
* the transaction because the inode cannot be unlocked once it is a
* part of the transaction.
*
* Start with zeroing any data block beyond EOF that we may expose on
* file extension.
*/
if (newsize > oldsize) {
error = xfs_zero_eof(ip, newsize, oldsize, &did_zeroing);
if (error)
return error;
}
/*
* We are going to log the inode size change in this transaction so
* any previous writes that are beyond the on disk EOF and the new
* EOF that have not been written out need to be written here. If we
* do not write the data out, we expose ourselves to the null files
* problem. Note that this includes any block zeroing we did above;
* otherwise those blocks may not be zeroed after a crash.
*/
if (newsize > ip->i_d.di_size &&
(oldsize != ip->i_d.di_size || did_zeroing)) {
error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
ip->i_d.di_size, newsize);
if (error)
return error;
}
/* Now wait for all direct I/O to complete. */
inode_dio_wait(inode);
/*
* We've already locked out new page faults, so now we can safely remove
* pages from the page cache knowing they won't get refaulted until we
* drop the XFS_MMAP_EXCL lock after the extent manipulations are
* complete. The truncate_setsize() call also cleans partial EOF page
* PTEs on extending truncates and hence ensures sub-page block size
* filesystems are correctly handled, too.
*
* We have to do all the page cache truncate work outside the
* transaction context as the "lock" order is page lock->log space
* reservation as defined by extent allocation in the writeback path.
* Hence a truncate can fail with ENOMEM from xfs_trans_reserve(), but
* having already truncated the in-memory version of the file (i.e. made
* user visible changes). There's not much we can do about this, except
* to hope that the caller sees ENOMEM and retries the truncate
* operation.
*/
if (IS_DAX(inode))
error = dax_truncate_page(inode, newsize, xfs_get_blocks_direct);
else
error = block_truncate_page(inode->i_mapping, newsize,
xfs_get_blocks);
if (error)
return error;
truncate_setsize(inode, newsize);
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
if (error)
goto out_trans_cancel;
lock_flags |= XFS_ILOCK_EXCL;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
/*
* Only change the c/mtime if we are changing the size or we are
* explicitly asked to change it. This handles the semantic difference
* between truncate() and ftruncate() as implemented in the VFS.
*
* The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
* special case where we need to update the times despite not having
* these flags set. For all other operations the VFS set these flags
* explicitly if it wants a timestamp update.
*/
if (newsize != oldsize &&
!(iattr->ia_valid & (ATTR_CTIME | ATTR_MTIME))) {
iattr->ia_ctime = iattr->ia_mtime =
current_fs_time(inode->i_sb);
iattr->ia_valid |= ATTR_CTIME | ATTR_MTIME;
}
/*
* The first thing we do is set the size to new_size permanently on
* disk. This way we don't have to worry about anyone ever being able
* to look at the data being freed even in the face of a crash.
* What we're getting around here is the case where we free a block, it
* is allocated to another file, it is written to, and then we crash.
* If the new data gets written to the file but the log buffers
* containing the free and reallocation don't, then we'd end up with
* garbage in the blocks being freed. As long as we make the new size
* permanent before actually freeing any blocks it doesn't matter if
* they get written to.
*/
ip->i_d.di_size = newsize;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (newsize <= oldsize) {
error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, newsize);
if (error)
goto out_trans_cancel;
/*
* Truncated "down", so we're removing references to old data
* here - if we delay flushing for a long time, we expose
* ourselves unduly to the notorious NULL files problem. So,
* we mark this inode and flush it when the file is closed,
* and do not wait the usual (long) time for writeout.
*/
xfs_iflags_set(ip, XFS_ITRUNCATED);
/* A truncate down always removes post-EOF blocks. */
xfs_inode_clear_eofblocks_tag(ip);
}
if (iattr->ia_valid & ATTR_MODE)
xfs_setattr_mode(ip, iattr);
if (iattr->ia_valid & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
xfs_setattr_time(ip, iattr);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(mp, xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp);
out_unlock:
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return error;
out_trans_cancel:
xfs_trans_cancel(tp);
goto out_unlock;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint resblks;
int committed;
int error;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN,
&ip->i_iocore, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
XFS_BMAP_INIT(&free_list, &firstfsb);
nimaps = 1;
error = XFS_BMAPI(mp, tp, io, offset_fsb, count_fsb,
XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
1, &imap, &nimaps, &free_list, NULL);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list),
firstfsb, &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES, NULL);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return XFS_ERROR(error);
if (unlikely(!imap.br_startblock &&
!(io->io_flags & XFS_IOCORE_RT)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap[XFS_STRAT_WRITE_IMAPS];
xfs_trans_t *tp;
int i, nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error == ENOSPC) {
error = xfs_trans_reserve(tp, 0,
XFS_WRITE_LOG_RES(mp),
0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
}
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
XFS_BMAP_INIT(&free_list, &first_block);
nimaps = XFS_STRAT_WRITE_IMAPS;
/*
* Ensure we don't go beyond eof - it is possible
* the extents changed since we did the read call,
* we dropped the ilock in the interim.
*/
end_fsb = XFS_B_TO_FSB(mp, ip->i_d.di_size);
xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = XFS_BMAPI(mp, tp, io, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
imap, &nimaps, &free_list, NULL);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list,
first_block, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp,
XFS_TRANS_RELEASE_LOG_RES, NULL);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
for (i = 0; i < nimaps; i++) {
if (unlikely(!imap[i].br_startblock &&
!(io->io_flags & XFS_IOCORE_RT)))
return xfs_cmn_err_fsblock_zero(ip, &imap[i]);
if ((offset_fsb >= imap[i].br_startoff) &&
(offset_fsb < (imap[i].br_startoff +
imap[i].br_blockcount))) {
*map = imap[i];
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
count_fsb -= imap[i].br_blockcount;
}
/* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
nimaps--;
map_start_fsb = imap[nimaps].br_startoff +
imap[nimaps].br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* Read in the ondisk dquot using dqtobp() then copy it to an incore version,
* and release the buffer immediately.
*
* If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed.
*/
int
xfs_qm_dqread(
struct xfs_mount *mp,
xfs_dqid_t id,
uint type,
uint flags,
struct xfs_dquot **O_dqpp)
{
struct xfs_dquot *dqp;
struct xfs_disk_dquot *ddqp;
struct xfs_buf *bp;
struct xfs_trans *tp = NULL;
int error;
int cancelflags = 0;
dqp = kmem_zone_zalloc(xfs_qm_dqzone, KM_SLEEP);
dqp->dq_flags = type;
dqp->q_core.d_id = cpu_to_be32(id);
dqp->q_mount = mp;
INIT_LIST_HEAD(&dqp->q_lru);
mutex_init(&dqp->q_qlock);
init_waitqueue_head(&dqp->q_pinwait);
/*
* Because we want to use a counting completion, complete
* the flush completion once to allow a single access to
* the flush completion without blocking.
*/
init_completion(&dqp->q_flush);
complete(&dqp->q_flush);
/*
* Make sure group quotas have a different lock class than user
* quotas.
*/
switch (type) {
case XFS_DQ_USER:
/* uses the default lock class */
break;
case XFS_DQ_GROUP:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_group_class);
break;
case XFS_DQ_PROJ:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_project_class);
break;
default:
ASSERT(0);
break;
}
XFS_STATS_INC(xs_qm_dquot);
trace_xfs_dqread(dqp);
if (flags & XFS_QMOPT_DQALLOC) {
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_dqalloc,
XFS_QM_DQALLOC_SPACE_RES(mp), 0);
if (error)
goto error1;
cancelflags = XFS_TRANS_RELEASE_LOG_RES;
}
/*
* get a pointer to the on-disk dquot and the buffer containing it
* dqp already knows its own type (GROUP/USER).
*/
error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags);
if (error) {
/*
* This can happen if quotas got turned off (ESRCH),
* or if the dquot didn't exist on disk and we ask to
* allocate (ENOENT).
*/
trace_xfs_dqread_fail(dqp);
cancelflags |= XFS_TRANS_ABORT;
goto error1;
}
/* copy everything from disk dquot to the incore dquot */
memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
xfs_qm_dquot_logitem_init(dqp);
/*
* Reservation counters are defined as reservation plus current usage
* to avoid having to add every time.
*/
dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount);
dqp->q_res_icount = be64_to_cpu(ddqp->d_icount);
dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount);
/* initialize the dquot speculative prealloc thresholds */
xfs_dquot_set_prealloc_limits(dqp);
/* Mark the buf so that this will stay incore a little longer */
xfs_buf_set_ref(bp, XFS_DQUOT_REF);
/*
* We got the buffer with a xfs_trans_read_buf() (in dqtobp())
* So we need to release with xfs_trans_brelse().
* The strategy here is identical to that of inodes; we lock
* the dquot in xfs_qm_dqget() before making it accessible to
* others. This is because dquots, like inodes, need a good level of
* concurrency, and we don't want to take locks on the entire buffers
* for dquot accesses.
* Note also that the dquot buffer may even be dirty at this point, if
* this particular dquot was repaired. We still aren't afraid to
* brelse it because we have the changes incore.
*/
ASSERT(xfs_buf_islocked(bp));
xfs_trans_brelse(tp, bp);
if (tp) {
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error0;
}
*O_dqpp = dqp;
return error;
error1:
if (tp)
xfs_trans_cancel(tp, cancelflags);
error0:
xfs_qm_dqdestroy(dqp);
*O_dqpp = NULL;
return error;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint resblks;
int committed;
int error;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, ip, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
/*
* Reserve enough blocks in this transaction for two complete extent
* btree splits. We may be converting the middle part of an unwritten
* extent and in this case we will insert two new extents in the btree
* each of which could cause a full split.
*
* This reservation amount will be used in the first call to
* xfs_bmbt_split() to select an AG with enough space to satisfy the
* rest of the operation.
*/
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*
* Note that we open code the transaction allocation here
* to pass KM_NOFS--we can't risk to recursing back into
* the filesystem here as we might be asked to write out
* the same inode that we complete here and might deadlock
* on the iolock.
*/
xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS);
tp->t_flags |= XFS_TRANS_RESERVE;
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
xfs_bmap_init(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
1, &imap, &nimaps, &free_list, NULL);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list), &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return XFS_ERROR(error);
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* xfs_rename
*/
int
xfs_rename(
xfs_inode_t *src_dp,
struct xfs_name *src_name,
xfs_inode_t *src_ip,
xfs_inode_t *target_dp,
struct xfs_name *target_name,
xfs_inode_t *target_ip)
{
xfs_trans_t *tp = NULL;
xfs_mount_t *mp = src_dp->i_mount;
int new_parent; /* moving to a new dir */
int src_is_directory; /* src_name is a directory */
int error;
xfs_bmap_free_t free_list;
xfs_fsblock_t first_block;
int cancel_flags;
int committed;
xfs_inode_t *inodes[4];
int spaceres;
int num_inodes;
xfs_itrace_entry(src_dp);
xfs_itrace_entry(target_dp);
if (DM_EVENT_ENABLED(src_dp, DM_EVENT_RENAME) ||
DM_EVENT_ENABLED(target_dp, DM_EVENT_RENAME)) {
error = XFS_SEND_NAMESP(mp, DM_EVENT_RENAME,
src_dp, DM_RIGHT_NULL,
target_dp, DM_RIGHT_NULL,
src_name->name, target_name->name,
0, 0, 0);
if (error)
return error;
}
/* Return through std_return after this point. */
new_parent = (src_dp != target_dp);
src_is_directory = ((src_ip->i_d.di_mode & S_IFMT) == S_IFDIR);
if (src_is_directory) {
/*
* Check for link count overflow on target_dp
*/
if (target_ip == NULL && new_parent &&
target_dp->i_d.di_nlink >= XFS_MAXLINK) {
error = XFS_ERROR(EMLINK);
goto std_return;
}
}
xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip,
inodes, &num_inodes);
xfs_bmap_init(&free_list, &first_block);
tp = xfs_trans_alloc(mp, XFS_TRANS_RENAME);
cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len);
error = xfs_trans_reserve(tp, spaceres, XFS_RENAME_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT);
if (error == ENOSPC) {
spaceres = 0;
error = xfs_trans_reserve(tp, 0, XFS_RENAME_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES, XFS_RENAME_LOG_COUNT);
}
if (error) {
xfs_trans_cancel(tp, 0);
goto std_return;
}
/*
* Attach the dquots to the inodes
*/
if ((error = XFS_QM_DQVOPRENAME(mp, inodes))) {
xfs_trans_cancel(tp, cancel_flags);
goto std_return;
}
/*
* Lock all the participating inodes. Depending upon whether
* the target_name exists in the target directory, and
* whether the target directory is the same as the source
* directory, we can lock from 2 to 4 inodes.
*/
xfs_lock_inodes(inodes, num_inodes, XFS_ILOCK_EXCL);
/*
* Join all the inodes to the transaction. From this point on,
* we can rely on either trans_commit or trans_cancel to unlock
* them. Note that we need to add a vnode reference to the
* directories since trans_commit & trans_cancel will decrement
* them when they unlock the inodes. Also, we need to be careful
* not to add an inode to the transaction more than once.
*/
IHOLD(src_dp);
xfs_trans_ijoin(tp, src_dp, XFS_ILOCK_EXCL);
if (new_parent) {
IHOLD(target_dp);
xfs_trans_ijoin(tp, target_dp, XFS_ILOCK_EXCL);
}
IHOLD(src_ip);
xfs_trans_ijoin(tp, src_ip, XFS_ILOCK_EXCL);
if (target_ip) {
IHOLD(target_ip);
xfs_trans_ijoin(tp, target_ip, XFS_ILOCK_EXCL);
}
/*
* If we are using project inheritance, we only allow renames
* into our tree when the project IDs are the same; else the
* tree quota mechanism would be circumvented.
*/
if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
(target_dp->i_d.di_projid != src_ip->i_d.di_projid))) {
error = XFS_ERROR(EXDEV);
goto error_return;
}
/*
* Set up the target.
*/
if (target_ip == NULL) {
/*
* If there's no space reservation, check the entry will
* fit before actually inserting it.
*/
error = xfs_dir_canenter(tp, target_dp, target_name, spaceres);
if (error)
goto error_return;
/*
* If target does not exist and the rename crosses
* directories, adjust the target directory link count
* to account for the ".." reference from the new entry.
*/
error = xfs_dir_createname(tp, target_dp, target_name,
src_ip->i_ino, &first_block,
&free_list, spaceres);
if (error == ENOSPC)
goto error_return;
if (error)
goto abort_return;
xfs_ichgtime(target_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
if (new_parent && src_is_directory) {
error = xfs_bumplink(tp, target_dp);
if (error)
goto abort_return;
}
} else { /* target_ip != NULL */
/*
* If target exists and it's a directory, check that both
* target and source are directories and that target can be
* destroyed, or that neither is a directory.
*/
if ((target_ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
/*
* Make sure target dir is empty.
*/
if (!(xfs_dir_isempty(target_ip)) ||
(target_ip->i_d.di_nlink > 2)) {
error = XFS_ERROR(EEXIST);
goto error_return;
}
}
/*
* Link the source inode under the target name.
* If the source inode is a directory and we are moving
* it across directories, its ".." entry will be
* inconsistent until we replace that down below.
*
* In case there is already an entry with the same
* name at the destination directory, remove it first.
*/
error = xfs_dir_replace(tp, target_dp, target_name,
src_ip->i_ino,
&first_block, &free_list, spaceres);
if (error)
goto abort_return;
xfs_ichgtime(target_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
/*
* Decrement the link count on the target since the target
* dir no longer points to it.
*/
error = xfs_droplink(tp, target_ip);
if (error)
goto abort_return;
if (src_is_directory) {
/*
* Drop the link from the old "." entry.
*/
error = xfs_droplink(tp, target_ip);
if (error)
goto abort_return;
}
} /* target_ip != NULL */
/*
* Remove the source.
*/
if (new_parent && src_is_directory) {
/*
* Rewrite the ".." entry to point to the new
* directory.
*/
error = xfs_dir_replace(tp, src_ip, &xfs_name_dotdot,
target_dp->i_ino,
&first_block, &free_list, spaceres);
ASSERT(error != EEXIST);
if (error)
goto abort_return;
}
/*
* We always want to hit the ctime on the source inode.
*
* This isn't strictly required by the standards since the source
* inode isn't really being changed, but old unix file systems did
* it and some incremental backup programs won't work without it.
*/
xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG);
/*
* Adjust the link count on src_dp. This is necessary when
* renaming a directory, either within one parent when
* the target existed, or across two parent directories.
*/
if (src_is_directory && (new_parent || target_ip != NULL)) {
/*
* Decrement link count on src_directory since the
* entry that's moved no longer points to it.
*/
error = xfs_droplink(tp, src_dp);
if (error)
goto abort_return;
}
error = xfs_dir_removename(tp, src_dp, src_name, src_ip->i_ino,
&first_block, &free_list, spaceres);
if (error)
goto abort_return;
xfs_ichgtime(src_dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
xfs_trans_log_inode(tp, src_dp, XFS_ILOG_CORE);
if (new_parent)
xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE);
/*
* If this is a synchronous mount, make sure that the
* rename transaction goes to disk before returning to
* the user.
*/
if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
xfs_trans_set_sync(tp);
}
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error) {
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES |
XFS_TRANS_ABORT));
goto std_return;
}
/*
* trans_commit will unlock src_ip, target_ip & decrement
* the vnode references.
*/
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
/* Fall through to std_return with error = 0 or errno from
* xfs_trans_commit */
std_return:
if (DM_EVENT_ENABLED(src_dp, DM_EVENT_POSTRENAME) ||
DM_EVENT_ENABLED(target_dp, DM_EVENT_POSTRENAME)) {
(void) XFS_SEND_NAMESP (mp, DM_EVENT_POSTRENAME,
src_dp, DM_RIGHT_NULL,
target_dp, DM_RIGHT_NULL,
src_name->name, target_name->name,
0, error, 0);
}
return error;
abort_return:
cancel_flags |= XFS_TRANS_ABORT;
/* FALLTHROUGH */
error_return:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, cancel_flags);
goto std_return;
}
/*
* Create an inode and return with a reference already taken, but unlocked
* This is how we create quota inodes
*/
STATIC int
xfs_qm_qino_alloc(
xfs_mount_t *mp,
xfs_inode_t **ip,
uint flags)
{
xfs_trans_t *tp;
int error;
int committed;
bool need_alloc = true;
*ip = NULL;
/*
* With superblock that doesn't have separate pquotino, we
* share an inode between gquota and pquota. If the on-disk
* superblock has GQUOTA and the filesystem is now mounted
* with PQUOTA, just use sb_gquotino for sb_pquotino and
* vice-versa.
*/
if (!xfs_sb_version_has_pquotino(&mp->m_sb) &&
(flags & (XFS_QMOPT_PQUOTA|XFS_QMOPT_GQUOTA))) {
xfs_ino_t ino = NULLFSINO;
if ((flags & XFS_QMOPT_PQUOTA) &&
(mp->m_sb.sb_gquotino != NULLFSINO)) {
ino = mp->m_sb.sb_gquotino;
ASSERT(mp->m_sb.sb_pquotino == NULLFSINO);
} else if ((flags & XFS_QMOPT_GQUOTA) &&
(mp->m_sb.sb_pquotino != NULLFSINO)) {
ino = mp->m_sb.sb_pquotino;
ASSERT(mp->m_sb.sb_gquotino == NULLFSINO);
}
if (ino != NULLFSINO) {
error = xfs_iget(mp, NULL, ino, 0, 0, ip);
if (error)
return error;
mp->m_sb.sb_gquotino = NULLFSINO;
mp->m_sb.sb_pquotino = NULLFSINO;
need_alloc = false;
}
}
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_create,
XFS_QM_QINOCREATE_SPACE_RES(mp), 0);
if (error) {
xfs_trans_cancel(tp, 0);
return error;
}
if (need_alloc) {
error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip,
&committed);
if (error) {
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
XFS_TRANS_ABORT);
return error;
}
}
/*
* Make the changes in the superblock, and log those too.
* sbfields arg may contain fields other than *QUOTINO;
* VERSIONNUM for example.
*/
spin_lock(&mp->m_sb_lock);
if (flags & XFS_QMOPT_SBVERSION) {
ASSERT(!xfs_sb_version_hasquota(&mp->m_sb));
xfs_sb_version_addquota(&mp->m_sb);
mp->m_sb.sb_uquotino = NULLFSINO;
mp->m_sb.sb_gquotino = NULLFSINO;
mp->m_sb.sb_pquotino = NULLFSINO;
/* qflags will get updated fully _after_ quotacheck */
mp->m_sb.sb_qflags = mp->m_qflags & XFS_ALL_QUOTA_ACCT;
}
if (flags & XFS_QMOPT_UQUOTA)
mp->m_sb.sb_uquotino = (*ip)->i_ino;
else if (flags & XFS_QMOPT_GQUOTA)
mp->m_sb.sb_gquotino = (*ip)->i_ino;
else
mp->m_sb.sb_pquotino = (*ip)->i_ino;
spin_unlock(&mp->m_sb_lock);
xfs_log_sb(tp);
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error) {
ASSERT(XFS_FORCED_SHUTDOWN(mp));
xfs_alert(mp, "%s failed (error %d)!", __func__, error);
}
if (need_alloc)
xfs_finish_inode_setup(*ip);
return error;
}
/*
* Truncate file. Must have write permission and not be a directory.
*/
int
xfs_setattr_size(
struct xfs_inode *ip,
struct iattr *iattr,
int flags)
{
struct xfs_mount *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
struct xfs_trans *tp;
int error;
uint lock_flags;
uint commit_flags = 0;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
ASSERT(S_ISREG(ip->i_d.di_mode));
ASSERT((mask & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID|
ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
lock_flags = XFS_ILOCK_EXCL;
if (!(flags & XFS_ATTR_NOLOCK))
lock_flags |= XFS_IOLOCK_EXCL;
xfs_ilock(ip, lock_flags);
/*
* Short circuit the truncate case for zero length files.
*/
if (iattr->ia_size == 0 &&
ip->i_size == 0 && ip->i_d.di_nextents == 0) {
if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
goto out_unlock;
/*
* Use the regular setattr path to update the timestamps.
*/
xfs_iunlock(ip, lock_flags);
iattr->ia_valid &= ~ATTR_SIZE;
return xfs_setattr_nonsize(ip, iattr, 0);
}
/*
* Make sure that the dquots are attached to the inode.
*/
error = xfs_qm_dqattach_locked(ip, 0);
if (error)
goto out_unlock;
/*
* Now we can make the changes. Before we join the inode to the
* transaction, take care of the part of the truncation that must be
* done without the inode lock. This needs to be done before joining
* the inode to the transaction, because the inode cannot be unlocked
* once it is a part of the transaction.
*/
if (iattr->ia_size > ip->i_size) {
/*
* Do the first part of growing a file: zero any data in the
* last block that is beyond the old EOF. We need to do this
* before the inode is joined to the transaction to modify
* i_size.
*/
error = xfs_zero_eof(ip, iattr->ia_size, ip->i_size);
if (error)
goto out_unlock;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
lock_flags &= ~XFS_ILOCK_EXCL;
/*
* We are going to log the inode size change in this transaction so
* any previous writes that are beyond the on disk EOF and the new
* EOF that have not been written out need to be written here. If we
* do not write the data out, we expose ourselves to the null files
* problem.
*
* Only flush from the on disk size to the smaller of the in memory
* file size or the new size as that's the range we really care about
* here and prevents waiting for other data not within the range we
* care about here.
*/
if (ip->i_size != ip->i_d.di_size && iattr->ia_size > ip->i_d.di_size) {
error = xfs_flush_pages(ip, ip->i_d.di_size, iattr->ia_size,
XBF_ASYNC, FI_NONE);
if (error)
goto out_unlock;
}
/*
* Wait for all I/O to complete.
*/
xfs_ioend_wait(ip);
error = -block_truncate_page(inode->i_mapping, iattr->ia_size,
xfs_get_blocks);
if (error)
goto out_unlock;
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_ITRUNCATE_LOG_COUNT);
if (error)
goto out_trans_cancel;
truncate_setsize(inode, iattr->ia_size);
commit_flags = XFS_TRANS_RELEASE_LOG_RES;
lock_flags |= XFS_ILOCK_EXCL;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip);
/*
* Only change the c/mtime if we are changing the size or we are
* explicitly asked to change it. This handles the semantic difference
* between truncate() and ftruncate() as implemented in the VFS.
*
* The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
* special case where we need to update the times despite not having
* these flags set. For all other operations the VFS set these flags
* explicitly if it wants a timestamp update.
*/
if (iattr->ia_size != ip->i_size &&
(!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
iattr->ia_ctime = iattr->ia_mtime =
current_fs_time(inode->i_sb);
mask |= ATTR_CTIME | ATTR_MTIME;
}
if (iattr->ia_size > ip->i_size) {
ip->i_d.di_size = iattr->ia_size;
ip->i_size = iattr->ia_size;
} else if (iattr->ia_size <= ip->i_size ||
(iattr->ia_size == 0 && ip->i_d.di_nextents)) {
error = xfs_itruncate_data(&tp, ip, iattr->ia_size);
if (error)
goto out_trans_abort;
/*
* Truncated "down", so we're removing references to old data
* here - if we delay flushing for a long time, we expose
* ourselves unduly to the notorious NULL files problem. So,
* we mark this inode and flush it when the file is closed,
* and do not wait the usual (long) time for writeout.
*/
xfs_iflags_set(ip, XFS_ITRUNCATED);
}
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
ip->i_update_core = 1;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
ip->i_update_core = 1;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
out_unlock:
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return error;
out_trans_abort:
commit_flags |= XFS_TRANS_ABORT;
out_trans_cancel:
xfs_trans_cancel(tp, commit_flags);
goto out_unlock;
}
int
xfs_attr_inactive(xfs_inode_t *dp)
{
xfs_trans_t *trans;
xfs_mount_t *mp;
int error;
mp = dp->i_mount;
ASSERT(! XFS_NOT_DQATTACHED(mp, dp));
xfs_ilock(dp, XFS_ILOCK_SHARED);
if (!xfs_inode_hasattr(dp) ||
dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
xfs_iunlock(dp, XFS_ILOCK_SHARED);
return 0;
}
xfs_iunlock(dp, XFS_ILOCK_SHARED);
/*
* Start our first transaction of the day.
*
* All future transactions during this code must be "chained" off
* this one via the trans_dup() call. All transactions will contain
* the inode, and the inode will always be marked with trans_ihold().
* Since the inode will be locked in all transactions, we must log
* the inode in every transaction to let it float upward through
* the log.
*/
trans = xfs_trans_alloc(mp, XFS_TRANS_ATTRINVAL);
error = xfs_trans_reserve(trans, &M_RES(mp)->tr_attrinval, 0, 0);
if (error) {
xfs_trans_cancel(trans, 0);
return error;
}
xfs_ilock(dp, XFS_ILOCK_EXCL);
/*
* No need to make quota reservations here. We expect to release some
* blocks, not allocate, in the common case.
*/
xfs_trans_ijoin(trans, dp, 0);
/*
* Decide on what work routines to call based on the inode size.
*/
if (!xfs_inode_hasattr(dp) ||
dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
error = 0;
goto out;
}
error = xfs_attr3_root_inactive(&trans, dp);
if (error)
goto out;
error = xfs_itruncate_extents(&trans, dp, XFS_ATTR_FORK, 0);
if (error)
goto out;
error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
out:
xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
}
static int
xfs_swap_extents(
xfs_inode_t *ip, /* target inode */
xfs_inode_t *tip, /* tmp inode */
xfs_swapext_t *sxp)
{
xfs_mount_t *mp = ip->i_mount;
xfs_trans_t *tp;
xfs_bstat_t *sbp = &sxp->sx_stat;
xfs_ifork_t *tempifp, *ifp, *tifp;
int src_log_flags, target_log_flags;
int error = 0;
int aforkblks = 0;
int taforkblks = 0;
__uint64_t tmp;
tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL);
if (!tempifp) {
error = XFS_ERROR(ENOMEM);
goto out;
}
/*
* we have to do two separate lock calls here to keep lockdep
* happy. If we try to get all the locks in one call, lock will
* report false positives when we drop the ILOCK and regain them
* below.
*/
xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
/* Verify that both files have the same format */
if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
/* Verify both files are either real-time or non-realtime */
if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
if (VN_CACHED(VFS_I(tip)) != 0) {
error = xfs_flushinval_pages(tip, 0, -1,
FI_REMAPF_LOCKED);
if (error)
goto out_unlock;
}
/* Verify O_DIRECT for ftmp */
if (VN_CACHED(VFS_I(tip)) != 0) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
/* Verify all data are being swapped */
if (sxp->sx_offset != 0 ||
sxp->sx_length != ip->i_d.di_size ||
sxp->sx_length != tip->i_d.di_size) {
error = XFS_ERROR(EFAULT);
goto out_unlock;
}
trace_xfs_swap_extent_before(ip, 0);
trace_xfs_swap_extent_before(tip, 1);
/* check inode formats now that data is flushed */
error = xfs_swap_extents_check_format(ip, tip);
if (error) {
xfs_notice(mp,
"%s: inode 0x%llx format is incompatible for exchanging.",
__func__, ip->i_ino);
goto out_unlock;
}
/*
* Compare the current change & modify times with that
* passed in. If they differ, we abort this swap.
* This is the mechanism used to ensure the calling
* process that the file was not changed out from
* under it.
*/
if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) ||
(sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
(sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
(sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
error = XFS_ERROR(EBUSY);
goto out_unlock;
}
/* We need to fail if the file is memory mapped. Once we have tossed
* all existing pages, the page fault will have no option
* but to go to the filesystem for pages. By making the page fault call
* vop_read (or write in the case of autogrow) they block on the iolock
* until we have switched the extents.
*/
if (VN_MAPPED(VFS_I(ip))) {
error = XFS_ERROR(EBUSY);
goto out_unlock;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_iunlock(tip, XFS_ILOCK_EXCL);
/*
* There is a race condition here since we gave up the
* ilock. However, the data fork will not change since
* we have the iolock (locked for truncation too) so we
* are safe. We don't really care if non-io related
* fields change.
*/
xfs_tosspages(ip, 0, -1, FI_REMAPF);
tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
if ((error = xfs_trans_reserve(tp, 0,
XFS_ICHANGE_LOG_RES(mp), 0,
0, 0))) {
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
xfs_iunlock(tip, XFS_IOLOCK_EXCL);
xfs_trans_cancel(tp, 0);
goto out;
}
xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
/*
* Count the number of extended attribute blocks
*/
if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) &&
(ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks);
if (error)
goto out_trans_cancel;
}
if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) &&
(tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK,
&taforkblks);
if (error)
goto out_trans_cancel;
}
/*
* Swap the data forks of the inodes
*/
ifp = &ip->i_df;
tifp = &tip->i_df;
*tempifp = *ifp; /* struct copy */
*ifp = *tifp; /* struct copy */
*tifp = *tempifp; /* struct copy */
/*
* Fix the on-disk inode values
*/
tmp = (__uint64_t)ip->i_d.di_nblocks;
ip->i_d.di_nblocks = tip->i_d.di_nblocks - taforkblks + aforkblks;
tip->i_d.di_nblocks = tmp + taforkblks - aforkblks;
tmp = (__uint64_t) ip->i_d.di_nextents;
ip->i_d.di_nextents = tip->i_d.di_nextents;
tip->i_d.di_nextents = tmp;
tmp = (__uint64_t) ip->i_d.di_format;
ip->i_d.di_format = tip->i_d.di_format;
tip->i_d.di_format = tmp;
/*
* The extents in the source inode could still contain speculative
* preallocation beyond EOF (e.g. the file is open but not modified
* while defrag is in progress). In that case, we need to copy over the
* number of delalloc blocks the data fork in the source inode is
* tracking beyond EOF so that when the fork is truncated away when the
* temporary inode is unlinked we don't underrun the i_delayed_blks
* counter on that inode.
*/
ASSERT(tip->i_delayed_blks == 0);
tip->i_delayed_blks = ip->i_delayed_blks;
ip->i_delayed_blks = 0;
src_log_flags = XFS_ILOG_CORE;
switch (ip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (ip->i_d.di_nextents <= XFS_INLINE_EXTS) {
ifp->if_u1.if_extents =
ifp->if_u2.if_inline_ext;
}
src_log_flags |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
src_log_flags |= XFS_ILOG_DBROOT;
break;
}
target_log_flags = XFS_ILOG_CORE;
switch (tip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (tip->i_d.di_nextents <= XFS_INLINE_EXTS) {
tifp->if_u1.if_extents =
tifp->if_u2.if_inline_ext;
}
target_log_flags |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
target_log_flags |= XFS_ILOG_DBROOT;
break;
}
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_log_inode(tp, ip, src_log_flags);
xfs_trans_log_inode(tp, tip, target_log_flags);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
trace_xfs_swap_extent_after(ip, 0);
trace_xfs_swap_extent_after(tip, 1);
out:
kmem_free(tempifp);
return error;
out_unlock:
xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
goto out;
out_trans_cancel:
xfs_trans_cancel(tp, 0);
goto out_unlock;
}
int
xfs_setattr_nonsize(
struct xfs_inode *ip,
struct iattr *iattr,
int flags)
{
xfs_mount_t *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
xfs_trans_t *tp;
int error;
uid_t uid = 0, iuid = 0;
gid_t gid = 0, igid = 0;
struct xfs_dquot *udqp = NULL, *gdqp = NULL;
struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
ASSERT((mask & ATTR_SIZE) == 0);
if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) {
uint qflags = 0;
if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) {
uid = iattr->ia_uid;
qflags |= XFS_QMOPT_UQUOTA;
} else {
uid = ip->i_d.di_uid;
}
if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) {
gid = iattr->ia_gid;
qflags |= XFS_QMOPT_GQUOTA;
} else {
gid = ip->i_d.di_gid;
}
ASSERT(udqp == NULL);
ASSERT(gdqp == NULL);
error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip),
qflags, &udqp, &gdqp);
if (error)
return error;
}
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
if (error)
goto out_dqrele;
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (mask & (ATTR_UID|ATTR_GID)) {
iuid = ip->i_d.di_uid;
igid = ip->i_d.di_gid;
gid = (mask & ATTR_GID) ? iattr->ia_gid : igid;
uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid;
if (XFS_IS_QUOTA_RUNNING(mp) &&
((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
(XFS_IS_GQUOTA_ON(mp) && igid != gid))) {
ASSERT(tp);
error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (error)
goto out_trans_cancel;
}
}
xfs_trans_ijoin(tp, ip, 0);
if (mask & (ATTR_UID|ATTR_GID)) {
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
if (iuid != uid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) {
ASSERT(mask & ATTR_UID);
ASSERT(udqp);
olddquot1 = xfs_qm_vop_chown(tp, ip,
&ip->i_udquot, udqp);
}
ip->i_d.di_uid = uid;
inode->i_uid = uid;
}
if (igid != gid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
ASSERT(!XFS_IS_PQUOTA_ON(mp));
ASSERT(mask & ATTR_GID);
ASSERT(gdqp);
olddquot2 = xfs_qm_vop_chown(tp, ip,
&ip->i_gdquot, gdqp);
}
ip->i_d.di_gid = gid;
inode->i_gid = gid;
}
}
if (mask & ATTR_MODE) {
umode_t mode = iattr->ia_mode;
if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
mode &= ~S_ISGID;
ip->i_d.di_mode &= S_IFMT;
ip->i_d.di_mode |= mode & ~S_IFMT;
inode->i_mode &= S_IFMT;
inode->i_mode |= mode & ~S_IFMT;
}
if (mask & ATTR_ATIME) {
inode->i_atime = iattr->ia_atime;
ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
}
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_qm_dqrele(olddquot1);
xfs_qm_dqrele(olddquot2);
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
if (error)
return XFS_ERROR(error);
if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
error = -xfs_acl_chmod(inode);
if (error)
return XFS_ERROR(error);
}
return 0;
out_trans_cancel:
xfs_trans_cancel(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
out_dqrele:
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
return error;
}
/*
* Create an inode and return with a reference already taken, but unlocked
* This is how we create quota inodes
*/
STATIC int
xfs_qm_qino_alloc(
xfs_mount_t *mp,
xfs_inode_t **ip,
__int64_t sbfields,
uint flags)
{
xfs_trans_t *tp;
int error;
int committed;
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE);
if ((error = xfs_trans_reserve(tp,
XFS_QM_QINOCREATE_SPACE_RES(mp),
XFS_CREATE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_CREATE_LOG_COUNT))) {
xfs_trans_cancel(tp, 0);
return error;
}
error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip, &committed);
if (error) {
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
XFS_TRANS_ABORT);
return error;
}
/*
* Make the changes in the superblock, and log those too.
* sbfields arg may contain fields other than *QUOTINO;
* VERSIONNUM for example.
*/
spin_lock(&mp->m_sb_lock);
if (flags & XFS_QMOPT_SBVERSION) {
ASSERT(!xfs_sb_version_hasquota(&mp->m_sb));
ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO |
XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) ==
(XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO |
XFS_SB_GQUOTINO | XFS_SB_QFLAGS));
xfs_sb_version_addquota(&mp->m_sb);
mp->m_sb.sb_uquotino = NULLFSINO;
mp->m_sb.sb_gquotino = NULLFSINO;
/* qflags will get updated _after_ quotacheck */
mp->m_sb.sb_qflags = 0;
}
if (flags & XFS_QMOPT_UQUOTA)
mp->m_sb.sb_uquotino = (*ip)->i_ino;
else
mp->m_sb.sb_gquotino = (*ip)->i_ino;
spin_unlock(&mp->m_sb_lock);
xfs_mod_sb(tp, sbfields);
if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) {
xfs_alert(mp, "%s failed (error %d)!", __func__, error);
return error;
}
return 0;
}
int
xfs_setattr_size(
struct xfs_inode *ip,
struct iattr *iattr,
int flags)
{
struct xfs_mount *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
xfs_off_t oldsize, newsize;
struct xfs_trans *tp;
int error;
uint lock_flags;
uint commit_flags = 0;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
ASSERT(S_ISREG(ip->i_d.di_mode));
ASSERT((mask & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID|
ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
lock_flags = XFS_ILOCK_EXCL;
if (!(flags & XFS_ATTR_NOLOCK))
lock_flags |= XFS_IOLOCK_EXCL;
xfs_ilock(ip, lock_flags);
oldsize = inode->i_size;
newsize = iattr->ia_size;
if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
goto out_unlock;
xfs_iunlock(ip, lock_flags);
iattr->ia_valid &= ~ATTR_SIZE;
return xfs_setattr_nonsize(ip, iattr, 0);
}
error = xfs_qm_dqattach_locked(ip, 0);
if (error)
goto out_unlock;
if (newsize > oldsize) {
error = xfs_zero_eof(ip, newsize, oldsize);
if (error)
goto out_unlock;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
lock_flags &= ~XFS_ILOCK_EXCL;
/*
* We are going to log the inode size change in this transaction so
* any previous writes that are beyond the on disk EOF and the new
* EOF that have not been written out need to be written here. If we
* do not write the data out, we expose ourselves to the null files
* problem.
*
* Only flush from the on disk size to the smaller of the in memory
* file size or the new size as that's the range we really care about
* here and prevents waiting for other data not within the range we
* care about here.
*/
if (oldsize != ip->i_d.di_size && newsize > ip->i_d.di_size) {
error = xfs_flush_pages(ip, ip->i_d.di_size, newsize, 0,
FI_NONE);
if (error)
goto out_unlock;
}
inode_dio_wait(inode);
error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
if (error)
goto out_unlock;
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_ITRUNCATE_LOG_COUNT);
if (error)
goto out_trans_cancel;
truncate_setsize(inode, newsize);
commit_flags = XFS_TRANS_RELEASE_LOG_RES;
lock_flags |= XFS_ILOCK_EXCL;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
iattr->ia_ctime = iattr->ia_mtime =
current_fs_time(inode->i_sb);
mask |= ATTR_CTIME | ATTR_MTIME;
}
/*
* The first thing we do is set the size to new_size permanently on
* disk. This way we don't have to worry about anyone ever being able
* to look at the data being freed even in the face of a crash.
* What we're getting around here is the case where we free a block, it
* is allocated to another file, it is written to, and then we crash.
* If the new data gets written to the file but the log buffers
* containing the free and reallocation don't, then we'd end up with
* garbage in the blocks being freed. As long as we make the new size
* permanent before actually freeing any blocks it doesn't matter if
* they get written to.
*/
ip->i_d.di_size = newsize;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (newsize <= oldsize) {
error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, newsize);
if (error)
goto out_trans_abort;
xfs_iflags_set(ip, XFS_ITRUNCATED);
}
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
out_unlock:
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return error;
out_trans_abort:
commit_flags |= XFS_TRANS_ABORT;
out_trans_cancel:
xfs_trans_cancel(tp, commit_flags);
goto out_unlock;
}
/*
* Adjust quota limits, and start/stop timers accordingly.
*/
STATIC int
xfs_qm_scall_setqlim(
xfs_mount_t *mp,
xfs_dqid_t id,
uint type,
fs_disk_quota_t *newlim)
{
xfs_disk_dquot_t *ddq;
xfs_dquot_t *dqp;
xfs_trans_t *tp;
int error;
xfs_qcnt_t hard, soft;
if (!capable(CAP_SYS_ADMIN))
return XFS_ERROR(EPERM);
if ((newlim->d_fieldmask & (FS_DQ_LIMIT_MASK|FS_DQ_TIMER_MASK)) == 0)
return (0);
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128,
0, 0, XFS_DEFAULT_LOG_COUNT))) {
xfs_trans_cancel(tp, 0);
return (error);
}
/*
* We don't want to race with a quotaoff so take the quotaoff lock.
* (We don't hold an inode lock, so there's nothing else to stop
* a quotaoff from happening). (XXXThis doesn't currently happen
* because we take the vfslock before calling xfs_qm_sysent).
*/
mutex_lock(&(XFS_QI_QOFFLOCK(mp)), PINOD);
/*
* Get the dquot (locked), and join it to the transaction.
* Allocate the dquot if this doesn't exist.
*/
if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) {
xfs_trans_cancel(tp, XFS_TRANS_ABORT);
mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
ASSERT(error != ENOENT);
return (error);
}
xfs_dqtrace_entry(dqp, "Q_SETQLIM: AFT DQGET");
xfs_trans_dqjoin(tp, dqp);
ddq = &dqp->q_core;
/*
* Make sure that hardlimits are >= soft limits before changing.
*/
hard = (newlim->d_fieldmask & FS_DQ_BHARD) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_hardlimit) :
INT_GET(ddq->d_blk_hardlimit, ARCH_CONVERT);
soft = (newlim->d_fieldmask & FS_DQ_BSOFT) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_softlimit) :
INT_GET(ddq->d_blk_softlimit, ARCH_CONVERT);
if (hard == 0 || hard >= soft) {
INT_SET(ddq->d_blk_hardlimit, ARCH_CONVERT, hard);
INT_SET(ddq->d_blk_softlimit, ARCH_CONVERT, soft);
if (id == 0) {
mp->m_quotainfo->qi_bhardlimit = hard;
mp->m_quotainfo->qi_bsoftlimit = soft;
}
} else {
qdprintk("blkhard %Ld < blksoft %Ld\n", hard, soft);
}
hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) :
INT_GET(ddq->d_rtb_hardlimit, ARCH_CONVERT);
soft = (newlim->d_fieldmask & FS_DQ_RTBSOFT) ?
(xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_softlimit) :
INT_GET(ddq->d_rtb_softlimit, ARCH_CONVERT);
if (hard == 0 || hard >= soft) {
INT_SET(ddq->d_rtb_hardlimit, ARCH_CONVERT, hard);
INT_SET(ddq->d_rtb_softlimit, ARCH_CONVERT, soft);
if (id == 0) {
mp->m_quotainfo->qi_rtbhardlimit = hard;
mp->m_quotainfo->qi_rtbsoftlimit = soft;
}
} else {
qdprintk("rtbhard %Ld < rtbsoft %Ld\n", hard, soft);
}
hard = (newlim->d_fieldmask & FS_DQ_IHARD) ?
(xfs_qcnt_t) newlim->d_ino_hardlimit :
INT_GET(ddq->d_ino_hardlimit, ARCH_CONVERT);
soft = (newlim->d_fieldmask & FS_DQ_ISOFT) ?
(xfs_qcnt_t) newlim->d_ino_softlimit :
INT_GET(ddq->d_ino_softlimit, ARCH_CONVERT);
if (hard == 0 || hard >= soft) {
INT_SET(ddq->d_ino_hardlimit, ARCH_CONVERT, hard);
INT_SET(ddq->d_ino_softlimit, ARCH_CONVERT, soft);
if (id == 0) {
mp->m_quotainfo->qi_ihardlimit = hard;
mp->m_quotainfo->qi_isoftlimit = soft;
}
} else {
qdprintk("ihard %Ld < isoft %Ld\n", hard, soft);
}
if (id == 0) {
/*
* Timelimits for the super user set the relative time
* the other users can be over quota for this file system.
* If it is zero a default is used. Ditto for the default
* soft and hard limit values (already done, above).
*/
if (newlim->d_fieldmask & FS_DQ_BTIMER) {
mp->m_quotainfo->qi_btimelimit = newlim->d_btimer;
INT_SET(ddq->d_btimer, ARCH_CONVERT, newlim->d_btimer);
}
if (newlim->d_fieldmask & FS_DQ_ITIMER) {
mp->m_quotainfo->qi_itimelimit = newlim->d_itimer;
INT_SET(ddq->d_itimer, ARCH_CONVERT, newlim->d_itimer);
}
if (newlim->d_fieldmask & FS_DQ_RTBTIMER) {
mp->m_quotainfo->qi_rtbtimelimit = newlim->d_rtbtimer;
INT_SET(ddq->d_rtbtimer, ARCH_CONVERT, newlim->d_rtbtimer);
}
} else /* if (XFS_IS_QUOTA_ENFORCED(mp)) */ {
/*
* If the user is now over quota, start the timelimit.
* The user will not be 'warned'.
* Note that we keep the timers ticking, whether enforcement
* is on or off. We don't really want to bother with iterating
* over all ondisk dquots and turning the timers on/off.
*/
xfs_qm_adjust_dqtimers(mp, ddq);
}
dqp->dq_flags |= XFS_DQ_DIRTY;
xfs_trans_log_dquot(tp, dqp);
xfs_dqtrace_entry(dqp, "Q_SETQLIM: COMMIT");
xfs_trans_commit(tp, 0, NULL);
xfs_qm_dqprint(dqp);
xfs_qm_dqrele(dqp);
mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
return (0);
}
STATIC long
xfs_file_fallocate(
struct file *file,
int mode,
loff_t offset,
loff_t len)
{
struct inode *inode = file_inode(file);
struct xfs_inode *ip = XFS_I(inode);
struct xfs_trans *tp;
long error;
loff_t new_size = 0;
if (!S_ISREG(inode->i_mode))
return -EINVAL;
if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE))
return -EOPNOTSUPP;
xfs_ilock(ip, XFS_IOLOCK_EXCL);
if (mode & FALLOC_FL_PUNCH_HOLE) {
error = xfs_free_file_space(ip, offset, len);
if (error)
goto out_unlock;
} else if (mode & FALLOC_FL_COLLAPSE_RANGE) {
unsigned blksize_mask = (1 << inode->i_blkbits) - 1;
if (offset & blksize_mask || len & blksize_mask) {
error = -EINVAL;
goto out_unlock;
}
ASSERT(offset + len < i_size_read(inode));
new_size = i_size_read(inode) - len;
error = xfs_collapse_file_space(ip, offset, len);
if (error)
goto out_unlock;
} else {
if (!(mode & FALLOC_FL_KEEP_SIZE) &&
offset + len > i_size_read(inode)) {
new_size = offset + len;
error = -inode_newsize_ok(inode, new_size);
if (error)
goto out_unlock;
}
if (mode & FALLOC_FL_ZERO_RANGE)
error = xfs_zero_file_space(ip, offset, len);
else
error = xfs_alloc_file_space(ip, offset, len,
XFS_BMAPI_PREALLOC);
if (error)
goto out_unlock;
}
tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_WRITEID);
error = xfs_trans_reserve(tp, &M_RES(ip->i_mount)->tr_writeid, 0, 0);
if (error) {
xfs_trans_cancel(tp, 0);
goto out_unlock;
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
ip->i_d.di_mode &= ~S_ISUID;
if (ip->i_d.di_mode & S_IXGRP)
ip->i_d.di_mode &= ~S_ISGID;
if (!(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE)))
ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (file->f_flags & O_DSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
if (error)
goto out_unlock;
/* Change file size if needed */
if (new_size) {
struct iattr iattr;
iattr.ia_valid = ATTR_SIZE;
iattr.ia_size = new_size;
error = xfs_setattr_size(ip, &iattr);
}
out_unlock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return -error;
}
/*
* Generic handler routine to remove a name from an attribute list.
* Transitions attribute list from Btree to shortform as necessary.
*/
int
xfs_attr_remove(
struct xfs_inode *dp,
const unsigned char *name,
int flags)
{
struct xfs_mount *mp = dp->i_mount;
struct xfs_da_args args;
struct xfs_bmap_free flist;
xfs_fsblock_t firstblock;
int error;
XFS_STATS_INC(xs_attr_remove);
if (XFS_FORCED_SHUTDOWN(dp->i_mount))
return -EIO;
if (!xfs_inode_hasattr(dp))
return -ENOATTR;
error = xfs_attr_args_init(&args, dp, name, flags);
if (error)
return error;
args.firstblock = &firstblock;
args.flist = &flist;
/*
* we have no control over the attribute names that userspace passes us
* to remove, so we have to allow the name lookup prior to attribute
* removal to fail.
*/
args.op_flags = XFS_DA_OP_OKNOENT;
error = xfs_qm_dqattach(dp, 0);
if (error)
return error;
/*
* Start our first transaction of the day.
*
* All future transactions during this code must be "chained" off
* this one via the trans_dup() call. All transactions will contain
* the inode, and the inode will always be marked with trans_ihold().
* Since the inode will be locked in all transactions, we must log
* the inode in every transaction to let it float upward through
* the log.
*/
args.trans = xfs_trans_alloc(mp, XFS_TRANS_ATTR_RM);
/*
* Root fork attributes can use reserved data blocks for this
* operation if necessary
*/
if (flags & ATTR_ROOT)
args.trans->t_flags |= XFS_TRANS_RESERVE;
error = xfs_trans_reserve(args.trans, &M_RES(mp)->tr_attrrm,
XFS_ATTRRM_SPACE_RES(mp), 0);
if (error) {
xfs_trans_cancel(args.trans, 0);
return error;
}
xfs_ilock(dp, XFS_ILOCK_EXCL);
/*
* No need to make quota reservations here. We expect to release some
* blocks not allocate in the common case.
*/
xfs_trans_ijoin(args.trans, dp, 0);
if (!xfs_inode_hasattr(dp)) {
error = -ENOATTR;
} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
ASSERT(dp->i_afp->if_flags & XFS_IFINLINE);
error = xfs_attr_shortform_remove(&args);
} else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
error = xfs_attr_leaf_removename(&args);
} else {
error = xfs_attr_node_removename(&args);
}
if (error)
goto out;
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(args.trans);
if ((flags & ATTR_KERNOTIME) == 0)
xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
/*
* Commit the last in the sequence of transactions.
*/
xfs_trans_log_inode(args.trans, dp, XFS_ILOG_CORE);
error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
out:
if (args.trans) {
xfs_trans_cancel(args.trans,
XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
}
xfs_iunlock(dp, XFS_ILOCK_EXCL);
return error;
}
int
xfs_ioc_space(
struct xfs_inode *ip,
struct inode *inode,
struct file *filp,
int ioflags,
unsigned int cmd,
xfs_flock64_t *bf)
{
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
struct iattr iattr;
bool setprealloc = false;
bool clrprealloc = false;
int error;
/*
* Only allow the sys admin to reserve space unless
* unwritten extents are enabled.
*/
if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) &&
!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
if (inode->i_flags & (S_IMMUTABLE|S_APPEND))
return -XFS_ERROR(EPERM);
if (!(filp->f_mode & FMODE_WRITE))
return -XFS_ERROR(EBADF);
if (!S_ISREG(inode->i_mode))
return -XFS_ERROR(EINVAL);
error = mnt_want_write_file(filp);
if (error)
return error;
xfs_ilock(ip, XFS_IOLOCK_EXCL);
switch (bf->l_whence) {
case 0: /*SEEK_SET*/
break;
case 1: /*SEEK_CUR*/
bf->l_start += filp->f_pos;
break;
case 2: /*SEEK_END*/
bf->l_start += XFS_ISIZE(ip);
break;
default:
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
/*
* length of <= 0 for resv/unresv/zero is invalid. length for
* alloc/free is ignored completely and we have no idea what userspace
* might have set it to, so set it to zero to allow range
* checks to pass.
*/
switch (cmd) {
case XFS_IOC_ZERO_RANGE:
case XFS_IOC_RESVSP:
case XFS_IOC_RESVSP64:
case XFS_IOC_UNRESVSP:
case XFS_IOC_UNRESVSP64:
if (bf->l_len <= 0) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
break;
default:
bf->l_len = 0;
break;
}
if (bf->l_start < 0 ||
bf->l_start > mp->m_super->s_maxbytes ||
bf->l_start + bf->l_len < 0 ||
bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
switch (cmd) {
case XFS_IOC_ZERO_RANGE:
error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
if (!error)
setprealloc = true;
break;
case XFS_IOC_RESVSP:
case XFS_IOC_RESVSP64:
error = xfs_alloc_file_space(ip, bf->l_start, bf->l_len,
XFS_BMAPI_PREALLOC);
if (!error)
setprealloc = true;
break;
case XFS_IOC_UNRESVSP:
case XFS_IOC_UNRESVSP64:
error = xfs_free_file_space(ip, bf->l_start, bf->l_len);
break;
case XFS_IOC_ALLOCSP:
case XFS_IOC_ALLOCSP64:
case XFS_IOC_FREESP:
case XFS_IOC_FREESP64:
if (bf->l_start > XFS_ISIZE(ip)) {
error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
bf->l_start - XFS_ISIZE(ip), 0);
if (error)
goto out_unlock;
}
iattr.ia_valid = ATTR_SIZE;
iattr.ia_size = bf->l_start;
error = xfs_setattr_size(ip, &iattr);
if (!error)
clrprealloc = true;
break;
default:
ASSERT(0);
error = XFS_ERROR(EINVAL);
}
if (error)
goto out_unlock;
tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_writeid, 0, 0);
if (error) {
xfs_trans_cancel(tp, 0);
goto out_unlock;
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
if (!(ioflags & IO_INVIS)) {
ip->i_d.di_mode &= ~S_ISUID;
if (ip->i_d.di_mode & S_IXGRP)
ip->i_d.di_mode &= ~S_ISGID;
xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
}
if (setprealloc)
ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
else if (clrprealloc)
ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (filp->f_flags & O_DSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
out_unlock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
mnt_drop_write_file(filp);
return -error;
}
STATIC int
xfs_ioctl_setattr(
xfs_inode_t *ip,
struct fsxattr *fa,
int mask)
{
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
unsigned int lock_flags = 0;
struct xfs_dquot *udqp = NULL;
struct xfs_dquot *gdqp = NULL;
struct xfs_dquot *olddquot = NULL;
int code;
trace_xfs_ioctl_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
/*
* Disallow 32bit project ids when projid32bit feature is not enabled.
*/
if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) &&
!xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb))
return XFS_ERROR(EINVAL);
/*
* If disk quotas is on, we make sure that the dquots do exist on disk,
* before we start any other transactions. Trying to do this later
* is messy. We don't care to take a readlock to look at the ids
* in inode here, because we can't hold it across the trans_reserve.
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) {
code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid,
ip->i_d.di_gid, fa->fsx_projid,
XFS_QMOPT_PQUOTA, &udqp, &gdqp);
if (code)
return code;
}
/*
* For the other attributes, we acquire the inode lock and
* first do an error checking pass.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
if (code)
goto error_return;
lock_flags = XFS_ILOCK_EXCL;
xfs_ilock(ip, lock_flags);
/*
* CAP_FOWNER overrides the following restrictions:
*
* The user ID of the calling process must be equal
* to the file owner ID, except in cases where the
* CAP_FSETID capability is applicable.
*/
if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
/*
* Do a quota reservation only if projid is actually going to change.
*/
if (mask & FSX_PROJID) {
if (XFS_IS_QUOTA_RUNNING(mp) &&
XFS_IS_PQUOTA_ON(mp) &&
xfs_get_projid(ip) != fa->fsx_projid) {
ASSERT(tp);
code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (code) /* out of quota */
goto error_return;
}
}
if (mask & FSX_EXTSIZE) {
/*
* Can't change extent size if any extents are allocated.
*/
if (ip->i_d.di_nextents &&
((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
fa->fsx_extsize)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* Extent size must be a multiple of the appropriate block
* size, if set at all. It must also be smaller than the
* maximum extent size supported by the filesystem.
*
* Also, for non-realtime files, limit the extent size hint to
* half the size of the AGs in the filesystem so alignment
* doesn't result in extents larger than an AG.
*/
if (fa->fsx_extsize != 0) {
xfs_extlen_t size;
xfs_fsblock_t extsize_fsb;
extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize);
if (extsize_fsb > MAXEXTLEN) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
if (XFS_IS_REALTIME_INODE(ip) ||
((mask & FSX_XFLAGS) &&
(fa->fsx_xflags & XFS_XFLAG_REALTIME))) {
size = mp->m_sb.sb_rextsize <<
mp->m_sb.sb_blocklog;
} else {
size = mp->m_sb.sb_blocksize;
if (extsize_fsb > mp->m_sb.sb_agblocks / 2) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
if (fa->fsx_extsize % size) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
}
if (mask & FSX_XFLAGS) {
/*
* Can't change realtime flag if any extents are allocated.
*/
if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
(XFS_IS_REALTIME_INODE(ip)) !=
(fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* If realtime flag is set then must have realtime data.
*/
if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
if ((mp->m_sb.sb_rblocks == 0) ||
(mp->m_sb.sb_rextsize == 0) ||
(ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
/*
* Can't modify an immutable/append-only file unless
* we have appropriate permission.
*/
if ((ip->i_d.di_flags &
(XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) ||
(fa->fsx_xflags &
(XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
!capable(CAP_LINUX_IMMUTABLE)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
}
xfs_trans_ijoin(tp, ip);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & FSX_PROJID) {
/*
* CAP_FSETID overrides the following restrictions:
*
* The set-user-ID and set-group-ID bits of a file will be
* cleared upon successful return from chown()
*/
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
/*
* Change the ownerships and register quota modifications
* in the transaction.
*/
if (xfs_get_projid(ip) != fa->fsx_projid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) {
olddquot = xfs_qm_vop_chown(tp, ip,
&ip->i_gdquot, gdqp);
}
xfs_set_projid(ip, fa->fsx_projid);
/*
* We may have to rev the inode as well as
* the superblock version number since projids didn't
* exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
*/
if (ip->i_d.di_version == 1)
xfs_bump_ino_vers2(tp, ip);
}
}
if (mask & FSX_EXTSIZE)
ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
if (mask & FSX_XFLAGS) {
xfs_set_diflags(ip, fa->fsx_xflags);
xfs_diflags_to_linux(ip);
}
xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
* This is slightly sub-optimal in that truncates require
* two sync transactions instead of one for wsync filesystems.
* One for the truncate and one for the timestamps since we
* don't want to change the timestamps unless we're sure the
* truncate worked. Truncates are less than 1% of the laddis
* mix so this probably isn't worth the trouble to optimize.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
code = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, lock_flags);
/*
* Release any dquot(s) the inode had kept before chown.
*/
xfs_qm_dqrele(olddquot);
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
return code;
error_return:
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
xfs_trans_cancel(tp, 0);
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return code;
}
/*
* Free a symlink that has blocks associated with it.
*/
STATIC int
xfs_inactive_symlink_rmt(
xfs_inode_t *ip,
xfs_trans_t **tpp)
{
xfs_buf_t *bp;
int committed;
int done;
int error;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
int i;
xfs_mount_t *mp;
xfs_bmbt_irec_t mval[XFS_SYMLINK_MAPS];
int nmaps;
xfs_trans_t *ntp;
int size;
xfs_trans_t *tp;
tp = *tpp;
mp = ip->i_mount;
ASSERT(ip->i_df.if_flags & XFS_IFEXTENTS);
/*
* We're freeing a symlink that has some
* blocks allocated to it. Free the
* blocks here. We know that we've got
* either 1 or 2 extents and that we can
* free them all in one bunmapi call.
*/
ASSERT(ip->i_d.di_nextents > 0 && ip->i_d.di_nextents <= 2);
/*
* Lock the inode, fix the size, and join it to the transaction.
* Hold it so in the normal path, we still have it locked for
* the second transaction. In the error paths we need it
* held so the cancel won't rele it, see below.
*/
size = (int)ip->i_d.di_size;
ip->i_d.di_size = 0;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
/*
* Find the block(s) so we can inval and unmap them.
*/
done = 0;
xfs_bmap_init(&free_list, &first_block);
nmaps = ARRAY_SIZE(mval);
error = xfs_bmapi_read(ip, 0, xfs_symlink_blocks(mp, size),
mval, &nmaps, 0);
if (error)
goto error0;
/*
* Invalidate the block(s). No validation is done.
*/
for (i = 0; i < nmaps; i++) {
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0);
if (!bp) {
error = ENOMEM;
goto error1;
}
xfs_trans_binval(tp, bp);
}
/*
* Unmap the dead block(s) to the free_list.
*/
if ((error = xfs_bunmapi(tp, ip, 0, size, XFS_BMAPI_METADATA, nmaps,
&first_block, &free_list, &done)))
goto error1;
ASSERT(done);
/*
* Commit the first transaction. This logs the EFI and the inode.
*/
if ((error = xfs_bmap_finish(&tp, &free_list, &committed)))
goto error1;
/*
* The transaction must have been committed, since there were
* actually extents freed by xfs_bunmapi. See xfs_bmap_finish.
* The new tp has the extent freeing and EFDs.
*/
ASSERT(committed);
/*
* The first xact was committed, so add the inode to the new one.
* Mark it dirty so it will be logged and moved forward in the log as
* part of every commit.
*/
xfs_trans_ijoin(tp, ip, 0);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
/*
* Get a new, empty transaction to return to our caller.
*/
ntp = xfs_trans_dup(tp);
/*
* Commit the transaction containing extent freeing and EFDs.
* If we get an error on the commit here or on the reserve below,
* we need to unlock the inode since the new transaction doesn't
* have the inode attached.
*/
error = xfs_trans_commit(tp, 0);
tp = ntp;
if (error) {
ASSERT(XFS_FORCED_SHUTDOWN(mp));
goto error0;
}
/*
* transaction commit worked ok so we can drop the extra ticket
* reference that we gained in xfs_trans_dup()
*/
xfs_log_ticket_put(tp->t_ticket);
/*
* Remove the memory for extent descriptions (just bookkeeping).
*/
if (ip->i_df.if_bytes)
xfs_idata_realloc(ip, -ip->i_df.if_bytes, XFS_DATA_FORK);
ASSERT(ip->i_df.if_bytes == 0);
/*
* Put an itruncate log reservation in the new transaction
* for our caller.
*/
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
if (error) {
ASSERT(XFS_FORCED_SHUTDOWN(mp));
goto error0;
}
xfs_trans_ijoin(tp, ip, 0);
*tpp = tp;
return 0;
error1:
xfs_bmap_cancel(&free_list);
error0:
return error;
}
