/*
* Clean a vnode of filesystem-specific data and prepare it for reuse.
*/
STATIC int
vn_reclaim(
struct vnode *vp)
{
int error;
XFS_STATS_INC(vn_reclaim);
vn_trace_entry(vp, "vn_reclaim", (inst_t *)__return_address);
/*
* Only make the VOP_RECLAIM call if there are behaviors
* to call.
*/
if (vp->v_fbhv) {
VOP_RECLAIM(vp, error);
if (error)
return -error;
}
ASSERT(vp->v_fbhv == NULL);
VN_LOCK(vp);
vp->v_flag &= (VRECLM|VWAIT);
VN_UNLOCK(vp, 0);
vp->v_type = VNON;
vp->v_fbhv = NULL;
#ifdef XFS_VNODE_TRACE
ktrace_free(vp->v_trace);
vp->v_trace = NULL;
#endif
return 0;
}
/*
* Look up an inode by number in the given file system.
* The inode is looked up in the cache held in each AG.
* If the inode is found in the cache, initialise the vfs inode
* if necessary.
*
* If it is not in core, read it in from the file system's device,
* add it to the cache and initialise the vfs inode.
*
* The inode is locked according to the value of the lock_flags parameter.
* This flag parameter indicates how and if the inode's IO lock and inode lock
* should be taken.
*
* mp -- the mount point structure for the current file system. It points
* to the inode hash table.
* tp -- a pointer to the current transaction if there is one. This is
* simply passed through to the xfs_iread() call.
* ino -- the number of the inode desired. This is the unique identifier
* within the file system for the inode being requested.
* lock_flags -- flags indicating how to lock the inode. See the comment
* for xfs_ilock() for a list of valid values.
*/
int
xfs_iget(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_ino_t ino,
uint flags,
uint lock_flags,
xfs_inode_t **ipp)
{
xfs_inode_t *ip;
int error;
xfs_perag_t *pag;
xfs_agino_t agino;
/* reject inode numbers outside existing AGs */
if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
return EINVAL;
/* get the perag structure and ensure that it's inode capable */
pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
agino = XFS_INO_TO_AGINO(mp, ino);
again:
error = 0;
rcu_read_lock();
ip = radix_tree_lookup(&pag->pag_ici_root, agino);
if (ip) {
error = xfs_iget_cache_hit(pag, ip, ino, flags, lock_flags);
if (error)
goto out_error_or_again;
} else {
rcu_read_unlock();
XFS_STATS_INC(xs_ig_missed);
error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip,
flags, lock_flags);
if (error)
goto out_error_or_again;
}
xfs_perag_put(pag);
*ipp = ip;
/*
* If we have a real type for an on-disk inode, we can set ops(&unlock)
* now. If it's a new inode being created, xfs_ialloc will handle it.
*/
if (xfs_iflags_test(ip, XFS_INEW) && ip->i_d.di_mode != 0)
xfs_setup_inode(ip);
return 0;
out_error_or_again:
if (error == EAGAIN) {
delay(1);
goto again;
}
xfs_perag_put(pag);
return error;
}
/*
* Read a directory.
*/
int
xfs_readdir(
xfs_inode_t *dp,
void *dirent,
size_t bufsize,
xfs_off_t *offset,
filldir_t filldir)
{
int rval; /* return value */
int v; /* type-checking value */
trace_xfs_readdir(dp);
if (XFS_FORCED_SHUTDOWN(dp->i_mount))
return XFS_ERROR(EIO);
ASSERT(S_ISDIR(dp->i_d.di_mode));
XFS_STATS_INC(xs_dir_getdents);
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_getdents(dp, dirent, offset, filldir);
else if ((rval = xfs_dir2_isblock(NULL, dp, &v)))
;
else if (v)
rval = xfs_dir2_block_getdents(dp, dirent, offset, filldir);
else
rval = xfs_dir2_leaf_getdents(dp, dirent, bufsize, offset,
filldir);
return rval;
}
STATIC ssize_t
xfs_file_read_iter(
struct kiocb *iocb,
struct iov_iter *to)
{
struct inode *inode = file_inode(iocb->ki_filp);
struct xfs_mount *mp = XFS_I(inode)->i_mount;
ssize_t ret = 0;
XFS_STATS_INC(mp, xs_read_calls);
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
if (IS_DAX(inode))
ret = xfs_file_dax_read(iocb, to);
else if (iocb->ki_flags & IOCB_DIRECT)
ret = xfs_file_dio_aio_read(iocb, to);
else
ret = xfs_file_buffered_aio_read(iocb, to);
if (ret > 0)
XFS_STATS_ADD(mp, xs_read_bytes, ret);
return ret;
}
STATIC ssize_t
xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
size_t count,
unsigned int flags)
{
struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
int ioflags = 0;
ssize_t ret;
XFS_STATS_INC(xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
STATIC void
xfs_qm_dqput_final(
struct xfs_dquot *dqp)
{
struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo;
struct xfs_dquot *gdqp;
trace_xfs_dqput_free(dqp);
mutex_lock(&qi->qi_lru_lock);
if (list_empty(&dqp->q_lru)) {
list_add_tail(&dqp->q_lru, &qi->qi_lru_list);
qi->qi_lru_count++;
XFS_STATS_INC(xs_qm_dquot_unused);
}
mutex_unlock(&qi->qi_lru_lock);
/*
* If we just added a udquot to the freelist, then we want to release
* the gdquot reference that it (probably) has. Otherwise it'll keep
* the gdquot from getting reclaimed.
*/
gdqp = dqp->q_gdquot;
if (gdqp) {
xfs_dqlock(gdqp);
dqp->q_gdquot = NULL;
}
xfs_dqunlock(dqp);
/*
* If we had a group quota hint, release it now.
*/
if (gdqp)
xfs_qm_dqput(gdqp);
}
/*
* Get a reference on a vnode.
*/
vnode_t *
vn_get(
struct vnode *vp,
vmap_t *vmap)
{
struct inode *inode;
XFS_STATS_INC(vn_get);
inode = LINVFS_GET_IP(vp);
if (inode->i_state & I_FREEING)
return NULL;
inode = VFS_GET_INODE(vmap->v_vfsp, vmap->v_ino, IGET_NOALLOC);
if (!inode) /* Inode not present */
return NULL;
/* We do not want to create new inodes via vn_get,
* returning NULL here is OK.
*/
if (inode->i_state & I_NEW) {
vn_mark_bad(vp);
unlock_new_inode(inode);
iput(inode);
return NULL;
}
vn_trace_exit(vp, "vn_get", (inst_t *)__return_address);
return vp;
}
int
xfs_attr_list_int(
xfs_attr_list_context_t *context)
{
int error;
xfs_inode_t *dp = context->dp;
uint lock_mode;
XFS_STATS_INC(dp->i_mount, xs_attr_list);
if (XFS_FORCED_SHUTDOWN(dp->i_mount))
return -EIO;
/*
* Decide on what work routines to call based on the inode size.
*/
lock_mode = xfs_ilock_attr_map_shared(dp);
if (!xfs_inode_hasattr(dp)) {
error = 0;
} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
error = xfs_attr_shortform_list(context);
} else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
error = xfs_attr_leaf_list(context);
} else {
error = xfs_attr_node_list(context);
}
xfs_iunlock(dp, lock_mode);
return error;
}
ssize_t
xfs_splice_write(
bhv_desc_t *bdp,
struct pipe_inode_info *pipe,
struct file *outfilp,
loff_t *ppos,
size_t count,
int flags,
int ioflags,
cred_t *credp)
{
xfs_inode_t *ip = XFS_BHVTOI(bdp);
xfs_mount_t *mp = ip->i_mount;
ssize_t ret;
struct inode *inode = outfilp->f_mapping->host;
xfs_fsize_t isize;
XFS_STATS_INC(xs_write_calls);
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_ilock(ip, XFS_IOLOCK_EXCL);
if (DM_EVENT_ENABLED(BHV_TO_VNODE(bdp)->v_vfsp, ip, DM_EVENT_WRITE) &&
(!(ioflags & IO_INVIS))) {
bhv_vrwlock_t locktype = VRWLOCK_WRITE;
int error;
error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, BHV_TO_VNODE(bdp),
*ppos, count,
FILP_DELAY_FLAG(outfilp), &locktype);
if (error) {
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return -error;
}
}
xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, &ip->i_iocore,
pipe, count, *ppos, ioflags);
ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
if (ret > 0)
XFS_STATS_ADD(xs_write_bytes, ret);
isize = i_size_read(inode);
if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
*ppos = isize;
if (*ppos > ip->i_d.di_size) {
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (*ppos > ip->i_d.di_size) {
ip->i_d.di_size = *ppos;
i_size_write(inode, *ppos);
ip->i_update_core = 1;
ip->i_update_size = 1;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
int
xfs_iget(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_ino_t ino,
uint flags,
uint lock_flags,
xfs_inode_t **ipp)
{
xfs_inode_t *ip;
int error;
xfs_perag_t *pag;
xfs_agino_t agino;
ASSERT((lock_flags & (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED)) == 0);
if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
return EINVAL;
pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
agino = XFS_INO_TO_AGINO(mp, ino);
again:
error = 0;
rcu_read_lock();
ip = radix_tree_lookup(&pag->pag_ici_root, agino);
if (ip) {
error = xfs_iget_cache_hit(pag, ip, ino, flags, lock_flags);
if (error)
goto out_error_or_again;
} else {
rcu_read_unlock();
XFS_STATS_INC(xs_ig_missed);
error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip,
flags, lock_flags);
if (error)
goto out_error_or_again;
}
xfs_perag_put(pag);
*ipp = ip;
if (xfs_iflags_test(ip, XFS_INEW) && ip->i_d.di_mode != 0)
xfs_setup_inode(ip);
return 0;
out_error_or_again:
if (error == EAGAIN) {
delay(1);
goto again;
}
xfs_perag_put(pag);
return error;
}
/*
* Remove an entry from a directory.
*/
int
xfs_dir_removename(
xfs_trans_t *tp,
xfs_inode_t *dp,
struct xfs_name *name,
xfs_ino_t ino,
xfs_fsblock_t *first, /* bmap's firstblock */
xfs_bmap_free_t *flist, /* bmap's freeblock list */
xfs_extlen_t total) /* bmap's total block count */
{
struct xfs_da_args *args;
int rval;
int v; /* type-checking value */
ASSERT(S_ISDIR(dp->i_d.di_mode));
XFS_STATS_INC(xs_dir_remove);
args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
if (!args)
return -ENOMEM;
args->geo = dp->i_mount->m_dir_geo;
args->name = name->name;
args->namelen = name->len;
args->filetype = name->type;
args->hashval = dp->i_mount->m_dirnameops->hashname(name);
args->inumber = ino;
args->dp = dp;
args->firstblock = first;
args->flist = flist;
args->total = total;
args->whichfork = XFS_DATA_FORK;
args->trans = tp;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
rval = xfs_dir2_sf_removename(args);
goto out_free;
}
rval = xfs_dir2_isblock(args, &v);
if (rval)
goto out_free;
if (v) {
rval = xfs_dir2_block_removename(args);
goto out_free;
}
rval = xfs_dir2_isleaf(args, &v);
if (rval)
goto out_free;
if (v)
rval = xfs_dir2_leaf_removename(args);
else
rval = xfs_dir2_node_removename(args);
out_free:
kmem_free(args);
return rval;
}
/*
* Implement readdir.
*/
static int /* error */
xfs_dir_getdents(xfs_trans_t *trans, xfs_inode_t *dp, uio_t *uio, int *eofp)
{
dirent_t *dbp;
int locklen = 0, alignment, retval, is32;
xfs_dir_put_t put;
int error;
XFS_STATS_INC(xs_dir_getdents);
ASSERT((dp->i_d.di_mode & IFMT) == IFDIR);
/*
* If our caller has given us a single contiguous memory buffer,
* just work directly within that buffer. If it's in user memory,
* lock it down first.
*/
is32 = ABI_IS_IRIX5(GETDENTS_ABI(get_current_abi(), uio));
alignment = (is32 ? sizeof(xfs32_off_t) : sizeof(xfs_off_t)) - 1;
if ((uio->uio_iovcnt == 1) &&
#if CELL_CAPABLE
!KT_CUR_ISXTHREAD() &&
#endif
(((__psint_t)uio->uio_iov[0].iov_base & alignment) == 0) &&
((uio->uio_iov[0].iov_len & alignment) == 0)) {
dbp = NULL;
if (uio->uio_segflag == UIO_SYSSPACE) {
put = xfs_dir_put_dirent64_direct;
} else {
put = is32 ?
xfs_dir_put_dirent32_direct :
xfs_dir_put_dirent64_direct;
}
} else {
dbp = kmem_alloc(sizeof(*dbp) + MAXNAMELEN, KM_SLEEP);
put = is32 ?
xfs_dir_put_dirent32_uio :
xfs_dir_put_dirent64_uio;
}
/*
* Decide on what work routines to call based on the inode size.
*/
*eofp = 0;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
retval = xfs_dir_shortform_getdents(dp, uio, eofp, dbp, put);
} else if (xfs_bmap_one_block(dp, XFS_DATA_FORK)) {
retval = xfs_dir_leaf_getdents(trans, dp, uio, eofp, dbp, put);
} else {
retval = xfs_dir_node_getdents(trans, dp, uio, eofp, dbp, put);
}
if (dbp != NULL)
kmem_free(dbp, sizeof(*dbp) + MAXNAMELEN);
return(retval);
}
/* Allocate and initialize everything we need for an incore dquot. */
STATIC struct xfs_dquot *
xfs_dquot_alloc(
struct xfs_mount *mp,
xfs_dqid_t id,
uint type)
{
struct xfs_dquot *dqp;
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);
dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
/*
* Offset of dquot in the (fixed sized) dquot chunk.
*/
dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
sizeof(xfs_dqblk_t);
/*
* 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_qm_dquot_logitem_init(dqp);
XFS_STATS_INC(mp, xs_qm_dquot);
return dqp;
}
ssize_t
xfs_sendfile(
bhv_desc_t *bdp,
struct file *filp,
loff_t *offset,
int ioflags,
size_t count,
read_actor_t actor,
void *target,
cred_t *credp)
{
ssize_t ret;
xfs_fsize_t n;
xfs_inode_t *ip;
xfs_mount_t *mp;
vnode_t *vp;
ip = XFS_BHVTOI(bdp);
vp = BHV_TO_VNODE(bdp);
mp = ip->i_mount;
vn_trace_entry(vp, "xfs_sendfile", (inst_t *)__return_address);
XFS_STATS_INC(xs_read_calls);
n = XFS_MAXIOFFSET(mp) - *offset;
if ((n <= 0) || (count == 0))
return 0;
if (n < count)
count = n;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_ilock(ip, XFS_IOLOCK_SHARED);
if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) &&
(!(ioflags & IO_INVIS))) {
vrwlock_t locktype = VRWLOCK_READ;
int error;
error = XFS_SEND_DATA(mp, DM_EVENT_READ, BHV_TO_VNODE(bdp), *offset, count,
FILP_DELAY_FLAG(filp), &locktype);
if (error) {
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
return -error;
}
}
ret = generic_file_sendfile(filp, offset, count, actor, target);
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_ichgtime(ip, XFS_ICHGTIME_ACC);
return ret;
}
STATIC ssize_t
xfs_file_aio_write(
struct kiocb *iocb,
const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
struct xfs_inode *ip = XFS_I(inode);
ssize_t ret;
size_t ocount = 0;
XFS_STATS_INC(xs_write_calls);
BUG_ON(iocb->ki_pos != pos);
ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
if (ret)
return ret;
if (ocount == 0)
return 0;
sb_start_write(inode->i_sb);
if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
ret = -EIO;
goto out;
}
if (unlikely(file->f_flags & O_DIRECT))
ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, ocount);
else
ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
ocount);
if (ret > 0) {
ssize_t err;
XFS_STATS_ADD(xs_write_bytes, ret);
/* Handle various SYNC-type writes */
err = generic_write_sync(file, pos, ret);
if (err < 0)
ret = err;
}
out:
sb_end_write(inode->i_sb);
return ret;
}
STATIC ssize_t
xfs_file_splice_from_socket(
struct file *file,
struct socket *sock,
loff_t __user *ppos,
size_t count)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
ssize_t ret;
xfs_fsize_t isize, new_size;
XFS_STATS_INC(xs_write_calls);
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_ilock(ip, XFS_IOLOCK_EXCL);
new_size = *ppos + count;
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (new_size > ip->i_size)
ip->i_new_size = new_size;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
// xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, ip,
// pipe, count, *ppos, ioflags);
ret = generic_splice_from_socket(file, sock, ppos, count);
if (ret > 0)
XFS_STATS_ADD(xs_write_bytes, ret);
isize = i_size_read(inode);
if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
*ppos = isize;
if (*ppos > ip->i_size) {
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (*ppos > ip->i_size)
ip->i_size = *ppos;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
if (ip->i_new_size) {
xfs_ilock(ip, XFS_ILOCK_EXCL);
ip->i_new_size = 0;
if (ip->i_d.di_size > ip->i_size)
ip->i_d.di_size = ip->i_size;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
/*
* Add a reference to a referenced vnode.
*/
bhv_vnode_t *
vn_hold(
bhv_vnode_t *vp)
{
struct inode *inode;
XFS_STATS_INC(vn_hold);
inode = igrab(vn_to_inode(vp));
ASSERT(inode);
return vp;
}
int
xfs_dir_lookup(
xfs_trans_t *tp,
xfs_inode_t *dp,
struct xfs_name *name,
xfs_ino_t *inum, /* out: inode number */
struct xfs_name *ci_name) /* out: actual name if CI match */
{
xfs_da_args_t args;
int rval;
int v; /* type-checking value */
ASSERT(S_ISDIR(dp->i_d.di_mode));
XFS_STATS_INC(xs_dir_lookup);
memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
args.filetype = name->type;
args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.dp = dp;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
args.op_flags = XFS_DA_OP_OKNOENT;
if (ci_name)
args.op_flags |= XFS_DA_OP_CILOOKUP;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_lookup(&args);
else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_block_lookup(&args);
else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_leaf_lookup(&args);
else
rval = xfs_dir2_node_lookup(&args);
if (rval == EEXIST)
rval = 0;
if (!rval) {
*inum = args.inumber;
if (ci_name) {
ci_name->name = args.value;
ci_name->len = args.valuelen;
}
}
return rval;
}
STATIC ssize_t
xfs_file_aio_write(
struct kiocb *iocb,
const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
struct xfs_inode *ip = XFS_I(inode);
ssize_t ret;
size_t ocount = 0;
XFS_STATS_INC(xs_write_calls);
BUG_ON(iocb->ki_pos != pos);
ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
if (ret)
return ret;
if (ocount == 0)
return 0;
xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE);
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
if (unlikely(file->f_flags & O_DIRECT))
ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, ocount);
else
ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
ocount);
if (ret > 0) {
ssize_t err;
XFS_STATS_ADD(xs_write_bytes, ret);
err = generic_write_sync(file, pos, ret);
if (err < 0)
ret = err;
}
return ret;
}
/*
* Add a reference to a referenced vnode.
*/
struct vnode *
vn_hold(
struct vnode *vp)
{
struct inode *inode;
XFS_STATS_INC(vn_hold);
VN_LOCK(vp);
inode = igrab(LINVFS_GET_IP(vp));
ASSERT(inode);
VN_UNLOCK(vp, 0);
return vp;
}
/*
Enter a name in a directory.
*/
int
xfs_dir_createname(
xfs_trans_t *tp,
xfs_inode_t *dp,
struct xfs_name *name,
xfs_ino_t inum, /* new entry inode number */
xfs_fsblock_t *first, /* bmap's firstblock */
xfs_bmap_free_t *flist, /* bmap's freeblock list */
xfs_extlen_t total) /* bmap's total block count */
{
xfs_da_args_t args;
int rval;
int v; /* type-checking value */
ASSERT(S_ISDIR(dp->i_d.di_mode));
if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
return rval;
XFS_STATS_INC(xs_dir_create);
memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
args.filetype = name->type;
args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.inumber = inum;
args.dp = dp;
args.firstblock = first;
args.flist = flist;
args.total = total;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_addname(&args);
else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_block_addname(&args);
else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_leaf_addname(&args);
else
rval = xfs_dir2_node_addname(&args);
return rval;
}
/*
Enter a name in a directory.
*/
int
xfs_dir_createname(
xfs_trans_t *tp,
xfs_inode_t *dp,
char *name,
int namelen,
xfs_ino_t inum, /* new entry inode number */
xfs_fsblock_t *first, /* bmap's firstblock */
xfs_bmap_free_t *flist, /* bmap's freeblock list */
xfs_extlen_t total) /* bmap's total block count */
{
xfs_da_args_t args;
int rval;
int v; /* type-checking value */
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
return rval;
XFS_STATS_INC(xs_dir_create);
args.name = name;
args.namelen = namelen;
args.hashval = xfs_da_hashname(name, namelen);
args.inumber = inum;
args.dp = dp;
args.firstblock = first;
args.flist = flist;
args.total = total;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
args.justcheck = 0;
args.addname = args.oknoent = 1;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_addname(&args);
else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_block_addname(&args);
else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_leaf_addname(&args);
else
rval = xfs_dir2_node_addname(&args);
return rval;
}
/*
* Lookup a name in a directory, give back the inode number.
*/
int
xfs_dir_lookup(
xfs_trans_t *tp,
xfs_inode_t *dp,
char *name,
int namelen,
xfs_ino_t *inum) /* out: inode number */
{
xfs_da_args_t args;
int rval;
int v; /* type-checking value */
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
XFS_STATS_INC(xs_dir_lookup);
args.name = name;
args.namelen = namelen;
args.hashval = xfs_da_hashname(name, namelen);
args.inumber = 0;
args.dp = dp;
args.firstblock = NULL;
args.flist = NULL;
args.total = 0;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
args.justcheck = args.addname = 0;
args.oknoent = 1;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_lookup(&args);
else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_block_lookup(&args);
else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
return rval;
else if (v)
rval = xfs_dir2_leaf_lookup(&args);
else
rval = xfs_dir2_node_lookup(&args);
if (rval == EEXIST)
rval = 0;
if (rval == 0)
*inum = args.inumber;
return rval;
}
/*
* Allocate and initialise an xfs_inode.
*/
struct xfs_inode *
xfs_inode_alloc(
struct xfs_mount *mp,
xfs_ino_t ino)
{
struct xfs_inode *ip;
/*
* if this didn't occur in transactions, we could use
* KM_MAYFAIL and return NULL here on ENOMEM. Set the
* code up to do this anyway.
*/
ip = kmem_zone_alloc(xfs_inode_zone, KM_SLEEP);
if (!ip)
return NULL;
if (inode_init_always(mp->m_super, VFS_I(ip))) {
kmem_zone_free(xfs_inode_zone, ip);
return NULL;
}
/* VFS doesn't initialise i_mode! */
VFS_I(ip)->i_mode = 0;
XFS_STATS_INC(mp, vn_active);
ASSERT(atomic_read(&ip->i_pincount) == 0);
ASSERT(!spin_is_locked(&ip->i_flags_lock));
ASSERT(!xfs_isiflocked(ip));
ASSERT(ip->i_ino == 0);
mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
/* initialise the xfs inode */
ip->i_ino = ino;
ip->i_mount = mp;
memset(&ip->i_imap, 0, sizeof(struct xfs_imap));
ip->i_afp = NULL;
ip->i_cowfp = NULL;
ip->i_cnextents = 0;
ip->i_cformat = XFS_DINODE_FMT_EXTENTS;
memset(&ip->i_df, 0, sizeof(xfs_ifork_t));
ip->i_flags = 0;
ip->i_delayed_blks = 0;
memset(&ip->i_d, 0, sizeof(ip->i_d));
return ip;
}
STATIC ssize_t
xfs_file_write_iter(
struct kiocb *iocb,
struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct address_space *mapping = file->f_mapping;
struct inode *inode = mapping->host;
struct xfs_inode *ip = XFS_I(inode);
ssize_t ret;
size_t ocount = iov_iter_count(from);
XFS_STATS_INC(ip->i_mount, xs_write_calls);
if (ocount == 0)
return 0;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
if (IS_DAX(inode))
ret = xfs_file_dax_write(iocb, from);
else if (iocb->ki_flags & IOCB_DIRECT) {
/*
* Allow a directio write to fall back to a buffered
* write *only* in the case that we're doing a reflink
* CoW. In all other directio scenarios we do not
* allow an operation to fall back to buffered mode.
*/
ret = xfs_file_dio_aio_write(iocb, from);
if (ret == -EREMCHG)
goto buffered;
} else {
buffered:
ret = xfs_file_buffered_aio_write(iocb, from);
}
if (ret > 0) {
XFS_STATS_ADD(ip->i_mount, xs_write_bytes, ret);
/* Handle various SYNC-type writes */
ret = generic_write_sync(iocb, ret);
}
return ret;
}
/*
* Call VOP_INACTIVE on last reference.
*/
void
vn_rele(
struct vnode *vp)
{
int vcnt;
int cache;
XFS_STATS_INC(vn_rele);
VN_LOCK(vp);
vn_trace_entry(vp, "vn_rele", (inst_t *)__return_address);
vcnt = vn_count(vp);
/*
* Since we always get called from put_inode we know
* that i_count won't be decremented after we
* return.
*/
if (!vcnt) {
/*
* As soon as we turn this on, noone can find us in vn_get
* until we turn off VINACT or VRECLM
*/
vp->v_flag |= VINACT;
VN_UNLOCK(vp, 0);
/*
* Do not make the VOP_INACTIVE call if there
* are no behaviors attached to the vnode to call.
*/
if (vp->v_fbhv)
VOP_INACTIVE(vp, NULL, cache);
VN_LOCK(vp);
if (vp->v_flag & VWAIT)
sv_broadcast(vptosync(vp));
vp->v_flag &= ~(VINACT|VWAIT|VRECLM|VMODIFIED);
}
VN_UNLOCK(vp, 0);
vn_trace_exit(vp, "vn_rele", (inst_t *)__return_address);
}
/*
* Release a reference to the dquot (decrement ref-count) and unlock it.
*
* If there is a group quota attached to this dquot, carefully release that
* too without tripping over deadlocks'n'stuff.
*/
void
xfs_qm_dqput(
struct xfs_dquot *dqp)
{
ASSERT(dqp->q_nrefs > 0);
ASSERT(XFS_DQ_IS_LOCKED(dqp));
trace_xfs_dqput(dqp);
if (--dqp->q_nrefs == 0) {
struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo;
trace_xfs_dqput_free(dqp);
if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
XFS_STATS_INC(xs_qm_dquot_unused);
}
xfs_dqunlock(dqp);
}
STATIC ssize_t
xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
size_t count,
unsigned int flags)
{
struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
struct xfs_mount *mp = ip->i_mount;
int ioflags = 0;
ssize_t ret;
XFS_STATS_INC(xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
xfs_ilock(ip, XFS_IOLOCK_SHARED);
if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
int iolock = XFS_IOLOCK_SHARED;
int error;
error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
FILP_DELAY_FLAG(infilp), &iolock);
if (error) {
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
return -error;
}
}
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
static int /* error */
xfs_dir_lookup(xfs_trans_t *trans, xfs_inode_t *dp, char *name, int namelen,
xfs_ino_t *inum)
{
xfs_da_args_t args;
int retval;
ASSERT((dp->i_d.di_mode & IFMT) == IFDIR);
if (namelen >= MAXNAMELEN) {
return(XFS_ERROR(EINVAL));
}
XFS_STATS_INC(xs_dir_lookup);
/*
* Fill in the arg structure for this request.
*/
args.name = name;
args.namelen = namelen;
args.hashval = xfs_da_hashname(name, namelen);
args.inumber = 0;
args.dp = dp;
args.firstblock = NULL;
args.flist = NULL;
args.total = 0;
args.whichfork = XFS_DATA_FORK;
args.trans = trans;
args.justcheck = args.addname = 0;
args.oknoent = 1;
/*
* Decide on what work routines to call based on the inode size.
*/
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
retval = xfs_dir_shortform_lookup(&args);
} else if (xfs_bmap_one_block(dp, XFS_DATA_FORK)) {
retval = xfs_dir_leaf_lookup(&args);
} else {
retval = xfs_dir_node_lookup(&args);
}
if (retval == EEXIST)
retval = 0;
*inum = args.inumber;
return(retval);
}
/*
* Read a directory.
*/
int
xfs_dir_getdents(
xfs_trans_t *tp,
xfs_inode_t *dp,
uio_t *uio, /* caller's buffer control */
int *eofp) /* out: eof reached */
{
int alignment; /* alignment required for ABI */
xfs_dirent_t *dbp; /* malloc'ed buffer */
xfs_dir2_put_t put; /* entry formatting routine */
int rval; /* return value */
int v; /* type-checking value */
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
XFS_STATS_INC(xs_dir_getdents);
/*
* If our caller has given us a single contiguous aligned memory buffer,
* just work directly within that buffer. If it's in user memory,
* lock it down first.
*/
alignment = sizeof(xfs_off_t) - 1;
if ((uio->uio_iovcnt == 1) &&
(((__psint_t)uio->uio_iov[0].iov_base & alignment) == 0) &&
((uio->uio_iov[0].iov_len & alignment) == 0)) {
dbp = NULL;
put = xfs_dir2_put_dirent64_direct;
} else {
dbp = kmem_alloc(sizeof(*dbp) + MAXNAMELEN, KM_SLEEP);
put = xfs_dir2_put_dirent64_uio;
}
*eofp = 0;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_getdents(dp, uio, eofp, dbp, put);
else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
;
else if (v)
rval = xfs_dir2_block_getdents(tp, dp, uio, eofp, dbp, put);
else
rval = xfs_dir2_leaf_getdents(tp, dp, uio, eofp, dbp, put);
if (dbp != NULL)
kmem_free(dbp, sizeof(*dbp) + MAXNAMELEN);
return rval;
}
