STATIC ssize_t
xfs_file_buffered_aio_write(
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;
int enospc = 0;
int iolock = XFS_IOLOCK_EXCL;
loff_t pos = iocb->ki_pos;
size_t count = iov_iter_count(from);
xfs_rw_ilock(ip, iolock);
ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
if (ret)
goto out;
iov_iter_truncate(from, count);
/* We can write back this queue in page reclaim */
current->backing_dev_info = mapping->backing_dev_info;
write_retry:
trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
ret = generic_perform_write(file, from, pos);
if (likely(ret >= 0))
iocb->ki_pos = pos + ret;
/*
* If we hit a space limit, try to free up some lingering preallocated
* space before returning an error. In the case of ENOSPC, first try to
* write back all dirty inodes to free up some of the excess reserved
* metadata space. This reduces the chances that the eofblocks scan
* waits on dirty mappings. Since xfs_flush_inodes() is serialized, this
* also behaves as a filter to prevent too many eofblocks scans from
* running at the same time.
*/
if (ret == -EDQUOT && !enospc) {
enospc = xfs_inode_free_quota_eofblocks(ip);
if (enospc)
goto write_retry;
} else if (ret == -ENOSPC && !enospc) {
struct xfs_eofblocks eofb = {0};
enospc = 1;
xfs_flush_inodes(ip->i_mount);
eofb.eof_scan_owner = ip->i_ino; /* for locking */
eofb.eof_flags = XFS_EOF_FLAGS_SYNC;
xfs_icache_free_eofblocks(ip->i_mount, &eofb);
goto write_retry;
}
current->backing_dev_info = NULL;
out:
xfs_rw_iunlock(ip, iolock);
return ret;
}
void
xfs_eofblocks_worker(
struct work_struct *work)
{
struct xfs_mount *mp = container_of(to_delayed_work(work),
struct xfs_mount, m_eofblocks_work);
xfs_icache_free_eofblocks(mp, NULL);
xfs_queue_eofblocks(mp);
}
/*
* Run eofblocks scans on the quotas applicable to the inode. For inodes with
* multiple quotas, we don't know exactly which quota caused an allocation
* failure. We make a best effort by including each quota under low free space
* conditions (less than 1% free space) in the scan.
*/
int
xfs_inode_free_quota_eofblocks(
struct xfs_inode *ip)
{
int scan = 0;
struct xfs_eofblocks eofb = {0};
struct xfs_dquot *dq;
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
/*
* Set the scan owner to avoid a potential livelock. Otherwise, the scan
* can repeatedly trylock on the inode we're currently processing. We
* run a sync scan to increase effectiveness and use the union filter to
* cover all applicable quotas in a single scan.
*/
eofb.eof_scan_owner = ip->i_ino;
eofb.eof_flags = XFS_EOF_FLAGS_UNION|XFS_EOF_FLAGS_SYNC;
if (XFS_IS_UQUOTA_ENFORCED(ip->i_mount)) {
dq = xfs_inode_dquot(ip, XFS_DQ_USER);
if (dq && xfs_dquot_lowsp(dq)) {
eofb.eof_uid = VFS_I(ip)->i_uid;
eofb.eof_flags |= XFS_EOF_FLAGS_UID;
scan = 1;
}
}
if (XFS_IS_GQUOTA_ENFORCED(ip->i_mount)) {
dq = xfs_inode_dquot(ip, XFS_DQ_GROUP);
if (dq && xfs_dquot_lowsp(dq)) {
eofb.eof_gid = VFS_I(ip)->i_gid;
eofb.eof_flags |= XFS_EOF_FLAGS_GID;
scan = 1;
}
}
if (scan)
xfs_icache_free_eofblocks(ip->i_mount, &eofb);
return scan;
}
/*
* Note: some of the ioctl's return positive numbers as a
* byte count indicating success, such as readlink_by_handle.
* So we don't "sign flip" like most other routines. This means
* true errors need to be returned as a negative value.
*/
long
xfs_file_ioctl(
struct file *filp,
unsigned int cmd,
unsigned long p)
{
struct inode *inode = file_inode(filp);
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
void __user *arg = (void __user *)p;
int ioflags = 0;
int error;
if (filp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
trace_xfs_file_ioctl(ip);
switch (cmd) {
case FITRIM:
return xfs_ioc_trim(mp, arg);
case XFS_IOC_ALLOCSP:
case XFS_IOC_FREESP:
case XFS_IOC_RESVSP:
case XFS_IOC_UNRESVSP:
case XFS_IOC_ALLOCSP64:
case XFS_IOC_FREESP64:
case XFS_IOC_RESVSP64:
case XFS_IOC_UNRESVSP64:
case XFS_IOC_ZERO_RANGE: {
xfs_flock64_t bf;
if (copy_from_user(&bf, arg, sizeof(bf)))
return -XFS_ERROR(EFAULT);
return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
}
case XFS_IOC_DIOINFO: {
struct dioattr da;
xfs_buftarg_t *target =
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp;
da.d_mem = da.d_miniosz = 1 << target->bt_sshift;
da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
if (copy_to_user(arg, &da, sizeof(da)))
return -XFS_ERROR(EFAULT);
return 0;
}
case XFS_IOC_FSBULKSTAT_SINGLE:
case XFS_IOC_FSBULKSTAT:
case XFS_IOC_FSINUMBERS:
return xfs_ioc_bulkstat(mp, cmd, arg);
case XFS_IOC_FSGEOMETRY_V1:
return xfs_ioc_fsgeometry_v1(mp, arg);
case XFS_IOC_FSGEOMETRY:
return xfs_ioc_fsgeometry(mp, arg);
case XFS_IOC_GETVERSION:
return put_user(inode->i_generation, (int __user *)arg);
case XFS_IOC_FSGETXATTR:
return xfs_ioc_fsgetxattr(ip, 0, arg);
case XFS_IOC_FSGETXATTRA:
return xfs_ioc_fsgetxattr(ip, 1, arg);
case XFS_IOC_FSSETXATTR:
return xfs_ioc_fssetxattr(ip, filp, arg);
case XFS_IOC_GETXFLAGS:
return xfs_ioc_getxflags(ip, arg);
case XFS_IOC_SETXFLAGS:
return xfs_ioc_setxflags(ip, filp, arg);
case XFS_IOC_FSSETDM: {
struct fsdmidata dmi;
if (copy_from_user(&dmi, arg, sizeof(dmi)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
dmi.fsd_dmstate);
mnt_drop_write_file(filp);
return -error;
}
case XFS_IOC_GETBMAP:
case XFS_IOC_GETBMAPA:
return xfs_ioc_getbmap(ip, ioflags, cmd, arg);
case XFS_IOC_GETBMAPX:
return xfs_ioc_getbmapx(ip, arg);
case XFS_IOC_FD_TO_HANDLE:
case XFS_IOC_PATH_TO_HANDLE:
case XFS_IOC_PATH_TO_FSHANDLE: {
xfs_fsop_handlereq_t hreq;
if (copy_from_user(&hreq, arg, sizeof(hreq)))
return -XFS_ERROR(EFAULT);
return xfs_find_handle(cmd, &hreq);
}
case XFS_IOC_OPEN_BY_HANDLE: {
xfs_fsop_handlereq_t hreq;
if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
return -XFS_ERROR(EFAULT);
return xfs_open_by_handle(filp, &hreq);
}
case XFS_IOC_FSSETDM_BY_HANDLE:
return xfs_fssetdm_by_handle(filp, arg);
case XFS_IOC_READLINK_BY_HANDLE: {
xfs_fsop_handlereq_t hreq;
if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
return -XFS_ERROR(EFAULT);
return xfs_readlink_by_handle(filp, &hreq);
}
case XFS_IOC_ATTRLIST_BY_HANDLE:
return xfs_attrlist_by_handle(filp, arg);
case XFS_IOC_ATTRMULTI_BY_HANDLE:
return xfs_attrmulti_by_handle(filp, arg);
case XFS_IOC_SWAPEXT: {
struct xfs_swapext sxp;
if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
error = xfs_swapext(&sxp);
mnt_drop_write_file(filp);
return -error;
}
case XFS_IOC_FSCOUNTS: {
xfs_fsop_counts_t out;
error = xfs_fs_counts(mp, &out);
if (error)
return -error;
if (copy_to_user(arg, &out, sizeof(out)))
return -XFS_ERROR(EFAULT);
return 0;
}
case XFS_IOC_SET_RESBLKS: {
xfs_fsop_resblks_t inout;
__uint64_t in;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (mp->m_flags & XFS_MOUNT_RDONLY)
return -XFS_ERROR(EROFS);
if (copy_from_user(&inout, arg, sizeof(inout)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
/* input parameter is passed in resblks field of structure */
in = inout.resblks;
error = xfs_reserve_blocks(mp, &in, &inout);
mnt_drop_write_file(filp);
if (error)
return -error;
if (copy_to_user(arg, &inout, sizeof(inout)))
return -XFS_ERROR(EFAULT);
return 0;
}
case XFS_IOC_GET_RESBLKS: {
xfs_fsop_resblks_t out;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
error = xfs_reserve_blocks(mp, NULL, &out);
if (error)
return -error;
if (copy_to_user(arg, &out, sizeof(out)))
return -XFS_ERROR(EFAULT);
return 0;
}
case XFS_IOC_FSGROWFSDATA: {
xfs_growfs_data_t in;
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
error = xfs_growfs_data(mp, &in);
mnt_drop_write_file(filp);
return -error;
}
case XFS_IOC_FSGROWFSLOG: {
xfs_growfs_log_t in;
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
error = xfs_growfs_log(mp, &in);
mnt_drop_write_file(filp);
return -error;
}
case XFS_IOC_FSGROWFSRT: {
xfs_growfs_rt_t in;
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
error = mnt_want_write_file(filp);
if (error)
return error;
error = xfs_growfs_rt(mp, &in);
mnt_drop_write_file(filp);
return -error;
}
case XFS_IOC_GOINGDOWN: {
__uint32_t in;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (get_user(in, (__uint32_t __user *)arg))
return -XFS_ERROR(EFAULT);
error = xfs_fs_goingdown(mp, in);
return -error;
}
case XFS_IOC_ERROR_INJECTION: {
xfs_error_injection_t in;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
error = xfs_errortag_add(in.errtag, mp);
return -error;
}
case XFS_IOC_ERROR_CLEARALL:
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
error = xfs_errortag_clearall(mp, 1);
return -error;
case XFS_IOC_FREE_EOFBLOCKS: {
struct xfs_eofblocks eofb;
if (copy_from_user(&eofb, arg, sizeof(eofb)))
return -XFS_ERROR(EFAULT);
if (eofb.eof_version != XFS_EOFBLOCKS_VERSION)
return -XFS_ERROR(EINVAL);
if (eofb.eof_flags & ~XFS_EOF_FLAGS_VALID)
return -XFS_ERROR(EINVAL);
if (memchr_inv(&eofb.pad32, 0, sizeof(eofb.pad32)) ||
memchr_inv(eofb.pad64, 0, sizeof(eofb.pad64)))
return -XFS_ERROR(EINVAL);
error = xfs_icache_free_eofblocks(mp, &eofb);
return -error;
}
default:
return -ENOTTY;
}
}
STATIC ssize_t
xfs_file_buffered_aio_write(
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;
int enospc = 0;
int iolock;
if (iocb->ki_flags & IOCB_NOWAIT)
return -EOPNOTSUPP;
write_retry:
iolock = XFS_IOLOCK_EXCL;
xfs_ilock(ip, iolock);
ret = xfs_file_aio_write_checks(iocb, from, &iolock);
if (ret)
goto out;
/* We can write back this queue in page reclaim */
current->backing_dev_info = inode_to_bdi(inode);
trace_xfs_file_buffered_write(ip, iov_iter_count(from), iocb->ki_pos);
ret = iomap_file_buffered_write(iocb, from, &xfs_iomap_ops);
if (likely(ret >= 0))
iocb->ki_pos += ret;
/*
* If we hit a space limit, try to free up some lingering preallocated
* space before returning an error. In the case of ENOSPC, first try to
* write back all dirty inodes to free up some of the excess reserved
* metadata space. This reduces the chances that the eofblocks scan
* waits on dirty mappings. Since xfs_flush_inodes() is serialized, this
* also behaves as a filter to prevent too many eofblocks scans from
* running at the same time.
*/
if (ret == -EDQUOT && !enospc) {
xfs_iunlock(ip, iolock);
enospc = xfs_inode_free_quota_eofblocks(ip);
if (enospc)
goto write_retry;
enospc = xfs_inode_free_quota_cowblocks(ip);
if (enospc)
goto write_retry;
iolock = 0;
} else if (ret == -ENOSPC && !enospc) {
struct xfs_eofblocks eofb = {0};
enospc = 1;
xfs_flush_inodes(ip->i_mount);
xfs_iunlock(ip, iolock);
eofb.eof_flags = XFS_EOF_FLAGS_SYNC;
xfs_icache_free_eofblocks(ip->i_mount, &eofb);
xfs_icache_free_cowblocks(ip->i_mount, &eofb);
goto write_retry;
}
current->backing_dev_info = NULL;
out:
if (iolock)
xfs_iunlock(ip, iolock);
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;
}