/*
* Hook in SELinux. This is not quite correct yet, what we really need
* here (as we do for default ACLs) is a mechanism by which creation of
* these attrs can be journalled at inode creation time (along with the
* inode, of course, such that log replay can't cause these to be lost).
*/
STATIC int
xfs_init_security(
bhv_vnode_t *vp,
struct inode *dir)
{
struct inode *ip = vn_to_inode(vp);
size_t length;
void *value;
char *name;
int error;
error = security_inode_init_security(ip, dir, &name, &value, &length);
if (error) {
if (error == -EOPNOTSUPP)
return 0;
return -error;
}
error = xfs_attr_set(XFS_I(ip), name, value,
length, ATTR_SECURE);
if (!error)
xfs_iflags_set(XFS_I(ip), XFS_IMODIFIED);
kfree(name);
kfree(value);
return error;
}
STATIC int
xfs_vn_symlink(
struct inode *dir,
struct dentry *dentry,
const char *symname)
{
struct inode *ip;
bhv_vnode_t *cvp; /* used to lookup symlink to put in dentry */
int error;
mode_t mode;
cvp = NULL;
mode = S_IFLNK |
(irix_symlink_mode ? 0777 & ~current->fs->umask : S_IRWXUGO);
error = xfs_symlink(XFS_I(dir), dentry, (char *)symname, mode,
&cvp, NULL);
if (likely(!error && cvp)) {
error = xfs_init_security(cvp, dir);
if (likely(!error)) {
ip = vn_to_inode(cvp);
d_instantiate(dentry, ip);
xfs_validate_fields(dir);
xfs_validate_fields(ip);
} else {
xfs_cleanup_inode(dir, cvp, dentry, 0);
}
}
return -error;
}
/*
* If the linux inode exists, mark it dirty.
* Used when commiting a dirty inode into a transaction so that
* the inode will get written back by the linux code
*/
void
xfs_mark_inode_dirty_sync(
xfs_inode_t *ip)
{
bhv_vnode_t *vp;
vp = XFS_ITOV_NULL(ip);
if (vp)
mark_inode_dirty_sync(vn_to_inode(vp));
}
/*
* 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;
}
/*
* Revalidate the Linux inode from the XFS inode.
* Note: i_size _not_ updated; we must hold the inode
* semaphore when doing that - callers responsibility.
*/
int
vn_revalidate(
bhv_vnode_t *vp)
{
struct inode *inode = vn_to_inode(vp);
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
unsigned long xflags;
xfs_itrace_entry(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
xfs_ilock(ip, XFS_ILOCK_SHARED);
inode->i_mode = ip->i_d.di_mode;
inode->i_uid = ip->i_d.di_uid;
inode->i_gid = ip->i_d.di_gid;
inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
xflags = xfs_ip2xflags(ip);
if (xflags & XFS_XFLAG_IMMUTABLE)
inode->i_flags |= S_IMMUTABLE;
else
inode->i_flags &= ~S_IMMUTABLE;
if (xflags & XFS_XFLAG_APPEND)
inode->i_flags |= S_APPEND;
else
inode->i_flags &= ~S_APPEND;
if (xflags & XFS_XFLAG_SYNC)
inode->i_flags |= S_SYNC;
else
inode->i_flags &= ~S_SYNC;
if (xflags & XFS_XFLAG_NOATIME)
inode->i_flags |= S_NOATIME;
else
inode->i_flags &= ~S_NOATIME;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
xfs_iflags_clear(ip, XFS_IMODIFIED);
return 0;
}
STATIC struct dentry *
xfs_fs_get_parent(
struct dentry *child)
{
int error;
bhv_vnode_t *cvp;
struct dentry *parent;
cvp = NULL;
error = xfs_lookup(XFS_I(child->d_inode), &dotdot, &cvp);
if (unlikely(error))
return ERR_PTR(-error);
parent = d_alloc_anon(vn_to_inode(cvp));
if (unlikely(!parent)) {
VN_RELE(cvp);
return ERR_PTR(-ENOMEM);
}
return parent;
}
STATIC struct dentry *
xfs_vn_lookup(
struct inode *dir,
struct dentry *dentry,
struct nameidata *nd)
{
bhv_vnode_t *cvp;
int error;
if (dentry->d_name.len >= MAXNAMELEN)
return ERR_PTR(-ENAMETOOLONG);
error = xfs_lookup(XFS_I(dir), dentry, &cvp);
if (unlikely(error)) {
if (unlikely(error != ENOENT))
return ERR_PTR(-error);
d_add(dentry, NULL);
return NULL;
}
return d_splice_alias(vn_to_inode(cvp), dentry);
}
/*
* Change the requested timestamp in the given inode.
* We don't lock across timestamp updates, and we don't log them but
* we do record the fact that there is dirty information in core.
*
* NOTE -- callers MUST combine XFS_ICHGTIME_MOD or XFS_ICHGTIME_CHG
* with XFS_ICHGTIME_ACC to be sure that access time
* update will take. Calling first with XFS_ICHGTIME_ACC
* and then XFS_ICHGTIME_MOD may fail to modify the access
* timestamp if the filesystem is mounted noacctm.
*/
void
xfs_ichgtime(
xfs_inode_t *ip,
int flags)
{
struct inode *inode = vn_to_inode(XFS_ITOV(ip));
timespec_t tv;
nanotime(&tv);
if (flags & XFS_ICHGTIME_MOD) {
inode->i_mtime = tv;
ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
}
if (flags & XFS_ICHGTIME_ACC) {
inode->i_atime = tv;
ip->i_d.di_atime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_atime.t_nsec = (__int32_t)tv.tv_nsec;
}
if (flags & XFS_ICHGTIME_CHG) {
inode->i_ctime = tv;
ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec;
}
/*
* We update the i_update_core field _after_ changing
* the timestamps in order to coordinate properly with
* xfs_iflush() so that we don't lose timestamp updates.
* This keeps us from having to hold the inode lock
* while doing this. We use the SYNCHRONIZE macro to
* ensure that the compiler does not reorder the update
* of i_update_core above the timestamp updates above.
*/
SYNCHRONIZE();
ip->i_update_core = 1;
if (!(inode->i_state & I_NEW))
mark_inode_dirty_sync(inode);
}
STATIC void
xfs_cleanup_inode(
struct inode *dir,
bhv_vnode_t *vp,
struct dentry *dentry,
int mode)
{
struct dentry teardown = {};
/* Oh, the horror.
* If we can't add the ACL or we fail in
* xfs_init_security we must back out.
* ENOSPC can hit here, among other things.
*/
teardown.d_inode = vn_to_inode(vp);
teardown.d_name = dentry->d_name;
if (S_ISDIR(mode))
xfs_rmdir(XFS_I(dir), &teardown);
else
xfs_remove(XFS_I(dir), &teardown);
VN_RELE(vp);
}
int /* error (positive) */
xfs_zero_eof(
vnode_t *vp,
xfs_iocore_t *io,
xfs_off_t offset, /* starting I/O offset */
xfs_fsize_t isize, /* current inode size */
xfs_fsize_t end_size) /* terminal inode size */
{
struct inode *ip = vn_to_inode(vp);
xfs_fileoff_t start_zero_fsb;
xfs_fileoff_t end_zero_fsb;
xfs_fileoff_t zero_count_fsb;
xfs_fileoff_t last_fsb;
xfs_extlen_t buf_len_fsb;
xfs_mount_t *mp = io->io_mount;
int nimaps;
int error = 0;
xfs_bmbt_irec_t imap;
ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
ASSERT(offset > isize);
/*
* First handle zeroing the block on which isize resides.
* We only zero a part of that block so it is handled specially.
*/
error = xfs_zero_last_block(ip, io, isize, end_size);
if (error) {
ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
return error;
}
/*
* Calculate the range between the new size and the old
* where blocks needing to be zeroed may exist. To get the
* block where the last byte in the file currently resides,
* we need to subtract one from the size and truncate back
* to a block boundary. We subtract 1 in case the size is
* exactly on a block boundary.
*/
last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
if (last_fsb == end_zero_fsb) {
/*
* The size was only incremented on its last block.
* We took care of that above, so just return.
*/
return 0;
}
ASSERT(start_zero_fsb <= end_zero_fsb);
while (start_zero_fsb <= end_zero_fsb) {
nimaps = 1;
zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
error = XFS_BMAPI(mp, NULL, io, start_zero_fsb, zero_count_fsb,
0, NULL, 0, &imap, &nimaps, NULL);
if (error) {
ASSERT(ismrlocked(io->io_lock, MR_UPDATE));
ASSERT(ismrlocked(io->io_iolock, MR_UPDATE));
return error;
}
ASSERT(nimaps > 0);
if (imap.br_state == XFS_EXT_UNWRITTEN ||
imap.br_startblock == HOLESTARTBLOCK) {
/*
* This loop handles initializing pages that were
* partially initialized by the code below this
* loop. It basically zeroes the part of the page
* that sits on a hole and sets the page as P_HOLE
* and calls remapf if it is a mapped file.
*/
start_zero_fsb = imap.br_startoff + imap.br_blockcount;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
continue;
}
/*
* There are blocks in the range requested.
* Zero them a single write at a time. We actually
* don't zero the entire range returned if it is
* too big and simply loop around to get the rest.
* That is not the most efficient thing to do, but it
* is simple and this path should not be exercised often.
*/
buf_len_fsb = XFS_FILBLKS_MIN(imap.br_blockcount,
mp->m_writeio_blocks << 8);
/*
* Drop the inode lock while we're doing the I/O.
* We'll still have the iolock to protect us.
*/
XFS_IUNLOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
error = xfs_iozero(ip,
XFS_FSB_TO_B(mp, start_zero_fsb),
XFS_FSB_TO_B(mp, buf_len_fsb),
end_size);
if (error) {
goto out_lock;
}
start_zero_fsb = imap.br_startoff + buf_len_fsb;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
}
return 0;
out_lock:
XFS_ILOCK(mp, io, XFS_ILOCK_EXCL|XFS_EXTSIZE_RD);
ASSERT(error >= 0);
return error;
}
/*
* Convert userspace handle data into vnode (and inode).
* We [ab]use the fact that all the fsop_handlereq ioctl calls
* have a data structure argument whose first component is always
* a xfs_fsop_handlereq_t, so we can cast to and from this type.
* This allows us to optimise the copy_from_user calls and gives
* a handy, shared routine.
*
* If no error, caller must always VN_RELE the returned vp.
*/
STATIC int
xfs_vget_fsop_handlereq(
xfs_mount_t *mp,
struct inode *parinode, /* parent inode pointer */
xfs_fsop_handlereq_t *hreq,
vnode_t **vp,
struct inode **inode)
{
void __user *hanp;
size_t hlen;
xfs_fid_t *xfid;
xfs_handle_t *handlep;
xfs_handle_t handle;
xfs_inode_t *ip;
struct inode *inodep;
vnode_t *vpp;
xfs_ino_t ino;
__u32 igen;
int error;
/*
* Only allow handle opens under a directory.
*/
if (!S_ISDIR(parinode->i_mode))
return XFS_ERROR(ENOTDIR);
hanp = hreq->ihandle;
hlen = hreq->ihandlen;
handlep = &handle;
if (hlen < sizeof(handlep->ha_fsid) || hlen > sizeof(*handlep))
return XFS_ERROR(EINVAL);
if (copy_from_user(handlep, hanp, hlen))
return XFS_ERROR(EFAULT);
if (hlen < sizeof(*handlep))
memset(((char *)handlep) + hlen, 0, sizeof(*handlep) - hlen);
if (hlen > sizeof(handlep->ha_fsid)) {
if (handlep->ha_fid.xfs_fid_len !=
(hlen - sizeof(handlep->ha_fsid)
- sizeof(handlep->ha_fid.xfs_fid_len))
|| handlep->ha_fid.xfs_fid_pad)
return XFS_ERROR(EINVAL);
}
/*
* Crack the handle, obtain the inode # & generation #
*/
xfid = (struct xfs_fid *)&handlep->ha_fid;
if (xfid->xfs_fid_len == sizeof(*xfid) - sizeof(xfid->xfs_fid_len)) {
ino = xfid->xfs_fid_ino;
igen = xfid->xfs_fid_gen;
} else {
return XFS_ERROR(EINVAL);
}
/*
* Get the XFS inode, building a vnode to go with it.
*/
error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_SHARED, &ip, 0);
if (error)
return error;
if (ip == NULL)
return XFS_ERROR(EIO);
if (ip->i_d.di_mode == 0 || ip->i_d.di_gen != igen) {
xfs_iput_new(ip, XFS_ILOCK_SHARED);
return XFS_ERROR(ENOENT);
}
vpp = XFS_ITOV(ip);
inodep = vn_to_inode(vpp);
xfs_iunlock(ip, XFS_ILOCK_SHARED);
*vp = vpp;
*inode = inodep;
return 0;
}
STATIC int
xfs_ioc_xattr(
vnode_t *vp,
xfs_inode_t *ip,
struct file *filp,
unsigned int cmd,
void __user *arg)
{
struct fsxattr fa;
struct vattr *vattr;
int error = 0;
int attr_flags;
unsigned int flags;
vattr = kmalloc(sizeof(*vattr), GFP_KERNEL);
if (unlikely(!vattr))
return -ENOMEM;
switch (cmd) {
case XFS_IOC_FSGETXATTR: {
vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
XFS_AT_NEXTENTS | XFS_AT_PROJID;
VOP_GETATTR(vp, vattr, 0, NULL, error);
if (unlikely(error)) {
error = -error;
break;
}
fa.fsx_xflags = vattr->va_xflags;
fa.fsx_extsize = vattr->va_extsize;
fa.fsx_nextents = vattr->va_nextents;
fa.fsx_projid = vattr->va_projid;
if (copy_to_user(arg, &fa, sizeof(fa))) {
error = -EFAULT;
break;
}
break;
}
case XFS_IOC_FSSETXATTR: {
if (copy_from_user(&fa, arg, sizeof(fa))) {
error = -EFAULT;
break;
}
attr_flags = 0;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
attr_flags |= ATTR_NONBLOCK;
vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | XFS_AT_PROJID;
vattr->va_xflags = fa.fsx_xflags;
vattr->va_extsize = fa.fsx_extsize;
vattr->va_projid = fa.fsx_projid;
VOP_SETATTR(vp, vattr, attr_flags, NULL, error);
if (likely(!error))
__vn_revalidate(vp, vattr); /* update flags */
error = -error;
break;
}
case XFS_IOC_FSGETXATTRA: {
vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
XFS_AT_ANEXTENTS | XFS_AT_PROJID;
VOP_GETATTR(vp, vattr, 0, NULL, error);
if (unlikely(error)) {
error = -error;
break;
}
fa.fsx_xflags = vattr->va_xflags;
fa.fsx_extsize = vattr->va_extsize;
fa.fsx_nextents = vattr->va_anextents;
fa.fsx_projid = vattr->va_projid;
if (copy_to_user(arg, &fa, sizeof(fa))) {
error = -EFAULT;
break;
}
break;
}
case XFS_IOC_GETXFLAGS: {
flags = xfs_di2lxflags(ip->i_d.di_flags);
if (copy_to_user(arg, &flags, sizeof(flags)))
error = -EFAULT;
break;
}
case XFS_IOC_SETXFLAGS: {
if (copy_from_user(&flags, arg, sizeof(flags))) {
error = -EFAULT;
break;
}
if (flags & ~(LINUX_XFLAG_IMMUTABLE | LINUX_XFLAG_APPEND | \
LINUX_XFLAG_NOATIME | LINUX_XFLAG_NODUMP | \
LINUX_XFLAG_SYNC)) {
error = -EOPNOTSUPP;
break;
}
attr_flags = 0;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
attr_flags |= ATTR_NONBLOCK;
vattr->va_mask = XFS_AT_XFLAGS;
vattr->va_xflags = xfs_merge_ioc_xflags(flags,
xfs_ip2xflags(ip));
VOP_SETATTR(vp, vattr, attr_flags, NULL, error);
if (likely(!error))
__vn_revalidate(vp, vattr); /* update flags */
error = -error;
break;
}
case XFS_IOC_GETVERSION: {
flags = vn_to_inode(vp)->i_generation;
if (copy_to_user(arg, &flags, sizeof(flags)))
error = -EFAULT;
break;
}
default:
error = -ENOTTY;
break;
}
kfree(vattr);
return error;
}
STATIC int
xfs_vn_mknod(
struct inode *dir,
struct dentry *dentry,
int mode,
dev_t rdev)
{
struct inode *ip;
bhv_vnode_t *vp = NULL, *dvp = vn_from_inode(dir);
xfs_acl_t *default_acl = NULL;
attrexists_t test_default_acl = _ACL_DEFAULT_EXISTS;
int error;
/*
* Irix uses Missed'em'V split, but doesn't want to see
* the upper 5 bits of (14bit) major.
*/
if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff))
return -EINVAL;
if (unlikely(test_default_acl && test_default_acl(dvp))) {
if (!_ACL_ALLOC(default_acl)) {
return -ENOMEM;
}
if (!_ACL_GET_DEFAULT(dvp, default_acl)) {
_ACL_FREE(default_acl);
default_acl = NULL;
}
}
if (IS_POSIXACL(dir) && !default_acl && xfs_has_fs_struct(current))
mode &= ~current->fs->umask;
switch (mode & S_IFMT) {
case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK:
rdev = sysv_encode_dev(rdev);
case S_IFREG:
error = xfs_create(XFS_I(dir), dentry, mode, rdev, &vp, NULL);
break;
case S_IFDIR:
error = xfs_mkdir(XFS_I(dir), dentry, mode, &vp, NULL);
break;
default:
error = EINVAL;
break;
}
if (unlikely(!error)) {
error = xfs_init_security(vp, dir);
if (error)
xfs_cleanup_inode(dir, vp, dentry, mode);
}
if (unlikely(default_acl)) {
if (!error) {
error = _ACL_INHERIT(vp, mode, default_acl);
if (!error)
xfs_iflags_set(XFS_I(vp), XFS_IMODIFIED);
else
xfs_cleanup_inode(dir, vp, dentry, mode);
}
_ACL_FREE(default_acl);
}
if (likely(!error)) {
ASSERT(vp);
ip = vn_to_inode(vp);
if (S_ISDIR(mode))
xfs_validate_fields(ip);
d_instantiate(dentry, ip);
xfs_validate_fields(dir);
}
return -error;
}
