void
xfs_alert_tag(
const struct xfs_mount *mp,
int panic_tag,
const char *fmt, ...)
{
struct va_format vaf;
va_list args;
int do_panic = 0;
if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) {
xfs_alert(mp, "Transforming an alert into a BUG.");
do_panic = 1;
}
va_start(args, fmt);
vaf.fmt = fmt;
vaf.va = &args;
__xfs_printk(KERN_ALERT, mp, &vaf);
va_end(args);
BUG_ON(do_panic);
}
/*
* Warnings specifically for verifier errors. Differentiate CRC vs. invalid
* values, and omit the stack trace unless the error level is tuned high.
*/
void
xfs_verifier_error(
struct xfs_buf *bp)
{
xfs_alert(NULL, "Metadata %s detected at %s block 0x%llx/0x%x",
bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
bp->b_ops->name, bp->b_bn, BBTOB(bp->b_length));
}
/*
* Force a shutdown of the filesystem instantly while keeping
* the filesystem consistent. We don't do an unmount here; just shutdown
* the shop, make sure that absolutely nothing persistent happens to
* this filesystem after this point.
*/
void
xfs_do_force_shutdown(
xfs_mount_t *mp,
int flags,
char *fname,
int lnnum)
{
int logerror;
logerror = flags & SHUTDOWN_LOG_IO_ERROR;
if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
xfs_notice(mp,
"%s(0x%x) called from line %d of file %s. Return address = 0x%p",
__func__, flags, lnnum, fname, __return_address);
}
/*
* No need to duplicate efforts.
*/
if (XFS_FORCED_SHUTDOWN(mp) && !logerror)
return;
/*
* This flags XFS_MOUNT_FS_SHUTDOWN, makes sure that we don't
* queue up anybody new on the log reservations, and wakes up
* everybody who's sleeping on log reservations to tell them
* the bad news.
*/
if (xfs_log_force_umount(mp, logerror))
return;
if (flags & SHUTDOWN_CORRUPT_INCORE) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_CORRUPT,
"Corruption of in-memory data detected. Shutting down filesystem");
if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
xfs_stack_trace();
} else if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
if (logerror) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_LOGERROR,
"Log I/O Error Detected. Shutting down filesystem");
} else if (flags & SHUTDOWN_DEVICE_REQ) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
"All device paths lost. Shutting down filesystem");
} else if (!(flags & SHUTDOWN_REMOTE_REQ)) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
"I/O Error Detected. Shutting down filesystem");
}
}
if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
xfs_alert(mp,
"Please umount the filesystem and rectify the problem(s)");
}
}
/*
* Prints out an ALERT message about I/O error.
*/
void
xfs_ioerror_alert(
char *func,
struct xfs_mount *mp,
xfs_buf_t *bp,
xfs_daddr_t blkno)
{
xfs_alert(mp,
"I/O error occurred: meta-data dev %s block 0x%llx"
" (\"%s\") error %d buf count %zd",
XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
(__uint64_t)blkno, func,
XFS_BUF_GETERROR(bp), XFS_BUF_COUNT(bp));
}
void
xfs_corruption_error(
const char *tag,
int level,
struct xfs_mount *mp,
void *p,
const char *filename,
int linenum,
inst_t *ra)
{
if (level <= xfs_error_level)
xfs_hex_dump(p, 64);
xfs_error_report(tag, level, mp, filename, linenum, ra);
xfs_alert(mp, "Corruption detected. Unmount and run xfs_repair");
}
/*
* If we are doing readahead on an inode buffer, we might be in log recovery
* reading an inode allocation buffer that hasn't yet been replayed, and hence
* has not had the inode cores stamped into it. Hence for readahead, the buffer
* may be potentially invalid.
*
* If the readahead buffer is invalid, we need to mark it with an error and
* clear the DONE status of the buffer so that a followup read will re-read it
* from disk. We don't report the error otherwise to avoid warnings during log
* recovery and we don't get unnecssary panics on debug kernels. We use EIO here
* because all we want to do is say readahead failed; there is no-one to report
* the error to, so this will distinguish it from a non-ra verifier failure.
* Changes to this readahead error behavour also need to be reflected in
* xfs_dquot_buf_readahead_verify().
*/
static void
xfs_inode_buf_verify(
struct xfs_buf *bp,
bool readahead)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
xfs_agnumber_t agno;
int i;
int ni;
/*
* Validate the magic number and version of every inode in the buffer
*/
agno = xfs_daddr_to_agno(mp, XFS_BUF_ADDR(bp));
ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
xfs_agino_t unlinked_ino;
dip = xfs_buf_offset(bp, (i << mp->m_sb.sb_inodelog));
unlinked_ino = be32_to_cpu(dip->di_next_unlinked);
di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) &&
xfs_dinode_good_version(mp, dip->di_version) &&
(unlinked_ino == NULLAGINO ||
xfs_verify_agino(mp, agno, unlinked_ino));
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP))) {
if (readahead) {
bp->b_flags &= ~XBF_DONE;
xfs_buf_ioerror(bp, -EIO);
return;
}
#ifdef DEBUG
xfs_alert(mp,
"bad inode magic/vsn daddr %lld #%d (magic=%x)",
(unsigned long long)bp->b_bn, i,
be16_to_cpu(dip->di_magic));
#endif
xfs_buf_verifier_error(bp, -EFSCORRUPTED,
__func__, dip, sizeof(*dip),
NULL);
return;
}
}
}
/*
* Helper functions to copy attribute data in and out of the one disk extents
*/
STATIC int
xfs_attr_rmtval_copyout(
struct xfs_mount *mp,
struct xfs_buf *bp,
xfs_ino_t ino,
int *offset,
int *valuelen,
__uint8_t **dst)
{
char *src = bp->b_addr;
xfs_daddr_t bno = bp->b_bn;
int len = BBTOB(bp->b_length);
int blksize = mp->m_attr_geo->blksize;
ASSERT(len >= blksize);
while (len > 0 && *valuelen > 0) {
int hdr_size = 0;
int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize);
byte_cnt = min(*valuelen, byte_cnt);
if (xfs_sb_version_hascrc(&mp->m_sb)) {
if (!xfs_attr3_rmt_hdr_ok(src, ino, *offset,
byte_cnt, bno)) {
xfs_alert(mp,
"remote attribute header mismatch bno/off/len/owner (0x%llx/0x%x/Ox%x/0x%llx)",
bno, *offset, byte_cnt, ino);
return -EFSCORRUPTED;
}
hdr_size = sizeof(struct xfs_attr3_rmt_hdr);
}
memcpy(*dst, src + hdr_size, byte_cnt);
/* roll buffer forwards */
len -= blksize;
src += blksize;
bno += BTOBB(blksize);
/* roll attribute data forwards */
*valuelen -= byte_cnt;
*dst += byte_cnt;
*offset += byte_cnt;
}
return 0;
}
/*
* xfs_inactive_symlink - free a symlink
*/
int
xfs_inactive_symlink(
struct xfs_inode *ip)
{
struct xfs_mount *mp = ip->i_mount;
int pathlen;
trace_xfs_inactive_symlink(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
xfs_ilock(ip, XFS_ILOCK_EXCL);
/*
* Zero length symlinks _can_ exist.
*/
pathlen = (int)ip->i_d.di_size;
if (!pathlen) {
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return 0;
}
if (pathlen < 0 || pathlen > MAXPATHLEN) {
xfs_alert(mp, "%s: inode (0x%llx) bad symlink length (%d)",
__func__, (unsigned long long)ip->i_ino, pathlen);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
ASSERT(0);
return XFS_ERROR(EFSCORRUPTED);
}
if (ip->i_df.if_flags & XFS_IFINLINE) {
if (ip->i_df.if_bytes > 0)
xfs_idata_realloc(ip, -(ip->i_df.if_bytes),
XFS_DATA_FORK);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
ASSERT(ip->i_df.if_bytes == 0);
return 0;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/* remove the remote symlink */
return xfs_inactive_symlink_rmt(ip);
}
void
xfs_do_force_shutdown(
xfs_mount_t *mp,
int flags,
char *fname,
int lnnum)
{
int logerror;
logerror = flags & SHUTDOWN_LOG_IO_ERROR;
if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
xfs_notice(mp,
"%s(0x%x) called from line %d of file %s. Return address = 0x%p",
__func__, flags, lnnum, fname, __return_address);
}
if (XFS_FORCED_SHUTDOWN(mp) && !logerror)
return;
if (xfs_log_force_umount(mp, logerror))
return;
if (flags & SHUTDOWN_CORRUPT_INCORE) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_CORRUPT,
"Corruption of in-memory data detected. Shutting down filesystem");
if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
xfs_stack_trace();
} else if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
if (logerror) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_LOGERROR,
"Log I/O Error Detected. Shutting down filesystem");
} else if (flags & SHUTDOWN_DEVICE_REQ) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
"All device paths lost. Shutting down filesystem");
} else if (!(flags & SHUTDOWN_REMOTE_REQ)) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
"I/O Error Detected. Shutting down filesystem");
}
}
if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
xfs_alert(mp,
"Please umount the filesystem and rectify the problem(s)");
}
}
void
xfs_inobp_check(
xfs_mount_t *mp,
xfs_buf_t *bp)
{
int i;
int j;
xfs_dinode_t *dip;
j = mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog;
for (i = 0; i < j; i++) {
dip = xfs_buf_offset(bp, i * mp->m_sb.sb_inodesize);
if (!dip->di_next_unlinked) {
xfs_alert(mp,
"Detected bogus zero next_unlinked field in inode %d buffer 0x%llx.",
i, (long long)bp->b_bn);
}
}
}
int
xfs_readlink(
struct xfs_inode *ip,
char *link)
{
struct xfs_mount *mp = ip->i_mount;
xfs_fsize_t pathlen;
int error = 0;
trace_xfs_readlink(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
xfs_ilock(ip, XFS_ILOCK_SHARED);
pathlen = ip->i_d.di_size;
if (!pathlen)
goto out;
if (pathlen < 0 || pathlen > MAXPATHLEN) {
xfs_alert(mp, "%s: inode (%llu) bad symlink length (%lld)",
__func__, (unsigned long long) ip->i_ino,
(long long) pathlen);
ASSERT(0);
error = XFS_ERROR(EFSCORRUPTED);
goto out;
}
if (ip->i_df.if_flags & XFS_IFINLINE) {
memcpy(link, ip->i_df.if_u1.if_data, pathlen);
link[pathlen] = '\0';
} else {
error = xfs_readlink_bmap(ip, link);
}
out:
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return error;
}
/*
* If we are doing readahead on an inode buffer, we might be in log recovery
* reading an inode allocation buffer that hasn't yet been replayed, and hence
* has not had the inode cores stamped into it. Hence for readahead, the buffer
* may be potentially invalid.
*
* If the readahead buffer is invalid, we don't want to mark it with an error,
* but we do want to clear the DONE status of the buffer so that a followup read
* will re-read it from disk. This will ensure that we don't get an unnecessary
* warnings during log recovery and we don't get unnecssary panics on debug
* kernels.
*/
static void
xfs_inode_buf_verify(
struct xfs_buf *bp,
bool readahead)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
int i;
int ni;
/*
* Validate the magic number and version of every inode in the buffer
*/
ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
dip = (struct xfs_dinode *)xfs_buf_offset(bp,
(i << mp->m_sb.sb_inodelog));
di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) &&
XFS_DINODE_GOOD_VERSION(dip->di_version);
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
if (readahead) {
bp->b_flags &= ~XBF_DONE;
return;
}
xfs_buf_ioerror(bp, -EFSCORRUPTED);
xfs_verifier_error(bp);
#ifdef DEBUG
xfs_alert(mp,
"bad inode magic/vsn daddr %lld #%d (magic=%x)",
(unsigned long long)bp->b_bn, i,
be16_to_cpu(dip->di_magic));
#endif
}
}
xfs_inobp_check(mp, bp);
}
int
xfs_readlink(
struct xfs_inode *ip,
char *link)
{
struct xfs_mount *mp = ip->i_mount;
xfs_fsize_t pathlen;
int error = 0;
trace_xfs_readlink(ip);
ASSERT(!(ip->i_df.if_flags & XFS_IFINLINE));
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
xfs_ilock(ip, XFS_ILOCK_SHARED);
pathlen = ip->i_d.di_size;
if (!pathlen)
goto out;
if (pathlen < 0 || pathlen > XFS_SYMLINK_MAXLEN) {
xfs_alert(mp, "%s: inode (%llu) bad symlink length (%lld)",
__func__, (unsigned long long) ip->i_ino,
(long long) pathlen);
ASSERT(0);
error = -EFSCORRUPTED;
goto out;
}
error = xfs_readlink_bmap_ilocked(ip, link);
out:
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return error;
}
/*
* This is called from xfs_mountfs to start quotas and initialize all
* necessary data structures like quotainfo. This is also responsible for
* running a quotacheck as necessary. We are guaranteed that the superblock
* is consistently read in at this point.
*
* If we fail here, the mount will continue with quota turned off. We don't
* need to inidicate success or failure at all.
*/
void
xfs_qm_mount_quotas(
xfs_mount_t *mp)
{
int error = 0;
uint sbf;
/*
* If quotas on realtime volumes is not supported, we disable
* quotas immediately.
*/
if (mp->m_sb.sb_rextents) {
xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
mp->m_qflags = 0;
goto write_changes;
}
ASSERT(XFS_IS_QUOTA_RUNNING(mp));
/*
* Allocate the quotainfo structure inside the mount struct, and
* create quotainode(s), and change/rev superblock if necessary.
*/
error = xfs_qm_init_quotainfo(mp);
if (error) {
/*
* We must turn off quotas.
*/
ASSERT(mp->m_quotainfo == NULL);
mp->m_qflags = 0;
goto write_changes;
}
/*
* If any of the quotas are not consistent, do a quotacheck.
*/
if (XFS_QM_NEED_QUOTACHECK(mp)) {
error = xfs_qm_quotacheck(mp);
if (error) {
/* Quotacheck failed and disabled quotas. */
return;
}
}
/*
* If one type of quotas is off, then it will lose its
* quotachecked status, since we won't be doing accounting for
* that type anymore.
*/
if (!XFS_IS_UQUOTA_ON(mp))
mp->m_qflags &= ~XFS_UQUOTA_CHKD;
if (!(XFS_IS_GQUOTA_ON(mp) || XFS_IS_PQUOTA_ON(mp)))
mp->m_qflags &= ~XFS_OQUOTA_CHKD;
write_changes:
/*
* We actually don't have to acquire the m_sb_lock at all.
* This can only be called from mount, and that's single threaded. XXX
*/
spin_lock(&mp->m_sb_lock);
sbf = mp->m_sb.sb_qflags;
mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL;
spin_unlock(&mp->m_sb_lock);
if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) {
if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) {
/*
* We could only have been turning quotas off.
* We aren't in very good shape actually because
* the incore structures are convinced that quotas are
* off, but the on disk superblock doesn't know that !
*/
ASSERT(!(XFS_IS_QUOTA_RUNNING(mp)));
xfs_alert(mp, "%s: Superblock update failed!",
__func__);
}
}
if (error) {
xfs_warn(mp, "Failed to initialize disk quotas.");
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;
}
/*
* Check the validity of the SB found.
*/
STATIC int
xfs_mount_validate_sb(
xfs_mount_t *mp,
xfs_sb_t *sbp,
bool check_inprogress,
bool check_version)
{
/*
* If the log device and data device have the
* same device number, the log is internal.
* Consequently, the sb_logstart should be non-zero. If
* we have a zero sb_logstart in this case, we may be trying to mount
* a volume filesystem in a non-volume manner.
*/
if (sbp->sb_magicnum != XFS_SB_MAGIC) {
xfs_warn(mp, "bad magic number");
return -EWRONGFS;
}
if (!xfs_sb_good_version(sbp)) {
xfs_warn(mp, "bad version");
return -EWRONGFS;
}
/*
* Version 5 superblock feature mask validation. Reject combinations the
* kernel cannot support up front before checking anything else. For
* write validation, we don't need to check feature masks.
*/
if (check_version && XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) {
if (xfs_sb_has_compat_feature(sbp,
XFS_SB_FEAT_COMPAT_UNKNOWN)) {
xfs_warn(mp,
"Superblock has unknown compatible features (0x%x) enabled.\n"
"Using a more recent kernel is recommended.",
(sbp->sb_features_compat &
XFS_SB_FEAT_COMPAT_UNKNOWN));
}
if (xfs_sb_has_ro_compat_feature(sbp,
XFS_SB_FEAT_RO_COMPAT_UNKNOWN)) {
xfs_alert(mp,
"Superblock has unknown read-only compatible features (0x%x) enabled.",
(sbp->sb_features_ro_compat &
XFS_SB_FEAT_RO_COMPAT_UNKNOWN));
if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
xfs_warn(mp,
"Attempted to mount read-only compatible filesystem read-write.\n"
"Filesystem can only be safely mounted read only.");
return -EINVAL;
}
}
if (xfs_sb_has_incompat_feature(sbp,
XFS_SB_FEAT_INCOMPAT_UNKNOWN)) {
xfs_warn(mp,
"Superblock has unknown incompatible features (0x%x) enabled.\n"
"Filesystem can not be safely mounted by this kernel.",
(sbp->sb_features_incompat &
XFS_SB_FEAT_INCOMPAT_UNKNOWN));
return -EINVAL;
}
}
if (xfs_sb_version_has_pquotino(sbp)) {
if (sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) {
xfs_notice(mp,
"Version 5 of Super block has XFS_OQUOTA bits.");
return -EFSCORRUPTED;
}
} else if (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD |
XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) {
xfs_notice(mp,
"Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.");
return -EFSCORRUPTED;
}
if (unlikely(
sbp->sb_logstart == 0 && mp->m_logdev_targp == mp->m_ddev_targp)) {
xfs_warn(mp,
"filesystem is marked as having an external log; "
"specify logdev on the mount command line.");
return -EINVAL;
}
if (unlikely(
sbp->sb_logstart != 0 && mp->m_logdev_targp != mp->m_ddev_targp)) {
xfs_warn(mp,
"filesystem is marked as having an internal log; "
"do not specify logdev on the mount command line.");
return -EINVAL;
}
/*
* More sanity checking. Most of these were stolen directly from
* xfs_repair.
*/
if (unlikely(
sbp->sb_agcount <= 0 ||
sbp->sb_sectsize < XFS_MIN_SECTORSIZE ||
sbp->sb_sectsize > XFS_MAX_SECTORSIZE ||
sbp->sb_sectlog < XFS_MIN_SECTORSIZE_LOG ||
sbp->sb_sectlog > XFS_MAX_SECTORSIZE_LOG ||
sbp->sb_sectsize != (1 << sbp->sb_sectlog) ||
sbp->sb_blocksize < XFS_MIN_BLOCKSIZE ||
sbp->sb_blocksize > XFS_MAX_BLOCKSIZE ||
sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG ||
sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG ||
sbp->sb_blocksize != (1 << sbp->sb_blocklog) ||
sbp->sb_inodesize < XFS_DINODE_MIN_SIZE ||
sbp->sb_inodesize > XFS_DINODE_MAX_SIZE ||
sbp->sb_inodelog < XFS_DINODE_MIN_LOG ||
sbp->sb_inodelog > XFS_DINODE_MAX_LOG ||
sbp->sb_inodesize != (1 << sbp->sb_inodelog) ||
sbp->sb_inopblock != howmany(sbp->sb_blocksize,sbp->sb_inodesize) ||
(sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog) ||
(sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE) ||
(sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) ||
(sbp->sb_imax_pct > 100 /* zero sb_imax_pct is valid */) ||
sbp->sb_dblocks == 0 ||
sbp->sb_dblocks > XFS_MAX_DBLOCKS(sbp) ||
sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp) ||
sbp->sb_shared_vn != 0)) {
xfs_notice(mp, "SB sanity check failed");
return -EFSCORRUPTED;
}
/*
* Until this is fixed only page-sized or smaller data blocks work.
*/
if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) {
xfs_warn(mp,
"File system with blocksize %d bytes. "
"Only pagesize (%ld) or less will currently work.",
sbp->sb_blocksize, PAGE_SIZE);
return -ENOSYS;
}
/*
* Currently only very few inode sizes are supported.
*/
switch (sbp->sb_inodesize) {
case 256:
case 512:
case 1024:
case 2048:
break;
default:
xfs_warn(mp, "inode size of %d bytes not supported",
sbp->sb_inodesize);
return -ENOSYS;
}
if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) ||
xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
xfs_warn(mp,
"file system too large to be mounted on this system.");
return -EFBIG;
}
if (check_inprogress && sbp->sb_inprogress) {
xfs_warn(mp, "Offline file system operation in progress!");
return -EFSCORRUPTED;
}
return 0;
}
/* ----- Kernel only functions below ----- */
STATIC int
xfs_readlink_bmap(
struct xfs_inode *ip,
char *link)
{
struct xfs_mount *mp = ip->i_mount;
struct xfs_bmbt_irec mval[XFS_SYMLINK_MAPS];
struct xfs_buf *bp;
xfs_daddr_t d;
char *cur_chunk;
int pathlen = ip->i_d.di_size;
int nmaps = XFS_SYMLINK_MAPS;
int byte_cnt;
int n;
int error = 0;
int fsblocks = 0;
int offset;
fsblocks = xfs_symlink_blocks(mp, pathlen);
error = xfs_bmapi_read(ip, 0, fsblocks, mval, &nmaps, 0);
if (error)
goto out;
offset = 0;
for (n = 0; n < nmaps; n++) {
d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
&xfs_symlink_buf_ops);
if (!bp)
return XFS_ERROR(ENOMEM);
error = bp->b_error;
if (error) {
xfs_buf_ioerror_alert(bp, __func__);
xfs_buf_relse(bp);
/* bad CRC means corrupted metadata */
if (error == EFSBADCRC)
error = EFSCORRUPTED;
goto out;
}
byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
if (pathlen < byte_cnt)
byte_cnt = pathlen;
cur_chunk = bp->b_addr;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
if (!xfs_symlink_hdr_ok(mp, ip->i_ino, offset,
byte_cnt, bp)) {
error = EFSCORRUPTED;
xfs_alert(mp,
"symlink header does not match required off/len/owner (0x%x/Ox%x,0x%llx)",
offset, byte_cnt, ip->i_ino);
xfs_buf_relse(bp);
goto out;
}
cur_chunk += sizeof(struct xfs_dsymlink_hdr);
}
memcpy(link + offset, bp->b_addr, byte_cnt);
pathlen -= byte_cnt;
offset += byte_cnt;
xfs_buf_relse(bp);
}
ASSERT(pathlen == 0);
link[ip->i_d.di_size] = '\0';
error = 0;
out:
return error;
}
/*
* Write a modified dquot to disk.
* The dquot must be locked and the flush lock too taken by caller.
* The flush lock will not be unlocked until the dquot reaches the disk,
* but the dquot is free to be unlocked and modified by the caller
* in the interim. Dquot is still locked on return. This behavior is
* identical to that of inodes.
*/
int
xfs_qm_dqflush(
struct xfs_dquot *dqp,
struct xfs_buf **bpp)
{
struct xfs_mount *mp = dqp->q_mount;
struct xfs_buf *bp;
struct xfs_dqblk *dqb;
struct xfs_disk_dquot *ddqp;
xfs_failaddr_t fa;
int error;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
ASSERT(!completion_done(&dqp->q_flush));
trace_xfs_dqflush(dqp);
*bpp = NULL;
xfs_qm_dqunpin_wait(dqp);
/*
* This may have been unpinned because the filesystem is shutting
* down forcibly. If that's the case we must not write this dquot
* to disk, because the log record didn't make it to disk.
*
* We also have to remove the log item from the AIL in this case,
* as we wait for an emptry AIL as part of the unmount process.
*/
if (XFS_FORCED_SHUTDOWN(mp)) {
struct xfs_log_item *lip = &dqp->q_logitem.qli_item;
dqp->dq_flags &= ~XFS_DQ_DIRTY;
xfs_trans_ail_remove(lip, SHUTDOWN_CORRUPT_INCORE);
error = -EIO;
goto out_unlock;
}
/*
* Get the buffer containing the on-disk dquot
*/
error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen, 0, &bp,
&xfs_dquot_buf_ops);
if (error)
goto out_unlock;
/*
* Calculate the location of the dquot inside the buffer.
*/
dqb = bp->b_addr + dqp->q_bufoffset;
ddqp = &dqb->dd_diskdq;
/*
* A simple sanity check in case we got a corrupted dquot.
*/
fa = xfs_dqblk_verify(mp, dqb, be32_to_cpu(ddqp->d_id), 0);
if (fa) {
xfs_alert(mp, "corrupt dquot ID 0x%x in memory at %pS",
be32_to_cpu(ddqp->d_id), fa);
xfs_buf_relse(bp);
xfs_dqfunlock(dqp);
xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
return -EIO;
}
/* This is the only portion of data that needs to persist */
memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t));
/*
* Clear the dirty field and remember the flush lsn for later use.
*/
dqp->dq_flags &= ~XFS_DQ_DIRTY;
xfs_trans_ail_copy_lsn(mp->m_ail, &dqp->q_logitem.qli_flush_lsn,
&dqp->q_logitem.qli_item.li_lsn);
/*
* copy the lsn into the on-disk dquot now while we have the in memory
* dquot here. This can't be done later in the write verifier as we
* can't get access to the log item at that point in time.
*
* We also calculate the CRC here so that the on-disk dquot in the
* buffer always has a valid CRC. This ensures there is no possibility
* of a dquot without an up-to-date CRC getting to disk.
*/
if (xfs_sb_version_hascrc(&mp->m_sb)) {
dqb->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn);
xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk),
XFS_DQUOT_CRC_OFF);
}
/*
* Attach an iodone routine so that we can remove this dquot from the
* AIL and release the flush lock once the dquot is synced to disk.
*/
xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done,
&dqp->q_logitem.qli_item);
/*
* If the buffer is pinned then push on the log so we won't
* get stuck waiting in the write for too long.
*/
if (xfs_buf_ispinned(bp)) {
trace_xfs_dqflush_force(dqp);
xfs_log_force(mp, 0);
}
trace_xfs_dqflush_done(dqp);
*bpp = bp;
return 0;
out_unlock:
xfs_dqfunlock(dqp);
return -EIO;
}
/*
* Do some primitive error checking on ondisk dquot data structures.
*/
int
xfs_dqcheck(
struct xfs_mount *mp,
xfs_disk_dquot_t *ddq,
xfs_dqid_t id,
uint type, /* used only when IO_dorepair is true */
uint flags,
char *str)
{
xfs_dqblk_t *d = (xfs_dqblk_t *)ddq;
int errs = 0;
/*
* We can encounter an uninitialized dquot buffer for 2 reasons:
* 1. If we crash while deleting the quotainode(s), and those blks got
* used for user data. This is because we take the path of regular
* file deletion; however, the size field of quotainodes is never
* updated, so all the tricks that we play in itruncate_finish
* don't quite matter.
*
* 2. We don't play the quota buffers when there's a quotaoff logitem.
* But the allocation will be replayed so we'll end up with an
* uninitialized quota block.
*
* This is all fine; things are still consistent, and we haven't lost
* any quota information. Just don't complain about bad dquot blks.
*/
if (ddq->d_magic != cpu_to_be16(XFS_DQUOT_MAGIC)) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : XFS dquot ID 0x%x, magic 0x%x != 0x%x",
str, id, be16_to_cpu(ddq->d_magic), XFS_DQUOT_MAGIC);
errs++;
}
if (ddq->d_version != XFS_DQUOT_VERSION) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : XFS dquot ID 0x%x, version 0x%x != 0x%x",
str, id, ddq->d_version, XFS_DQUOT_VERSION);
errs++;
}
if (ddq->d_flags != XFS_DQ_USER &&
ddq->d_flags != XFS_DQ_PROJ &&
ddq->d_flags != XFS_DQ_GROUP) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : XFS dquot ID 0x%x, unknown flags 0x%x",
str, id, ddq->d_flags);
errs++;
}
if (id != -1 && id != be32_to_cpu(ddq->d_id)) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : ondisk-dquot 0x%p, ID mismatch: "
"0x%x expected, found id 0x%x",
str, ddq, id, be32_to_cpu(ddq->d_id));
errs++;
}
if (!errs && ddq->d_id) {
if (ddq->d_blk_softlimit &&
be64_to_cpu(ddq->d_bcount) >
be64_to_cpu(ddq->d_blk_softlimit)) {
if (!ddq->d_btimer) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : Dquot ID 0x%x (0x%p) BLK TIMER NOT STARTED",
str, (int)be32_to_cpu(ddq->d_id), ddq);
errs++;
}
}
if (ddq->d_ino_softlimit &&
be64_to_cpu(ddq->d_icount) >
be64_to_cpu(ddq->d_ino_softlimit)) {
if (!ddq->d_itimer) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : Dquot ID 0x%x (0x%p) INODE TIMER NOT STARTED",
str, (int)be32_to_cpu(ddq->d_id), ddq);
errs++;
}
}
if (ddq->d_rtb_softlimit &&
be64_to_cpu(ddq->d_rtbcount) >
be64_to_cpu(ddq->d_rtb_softlimit)) {
if (!ddq->d_rtbtimer) {
if (flags & XFS_QMOPT_DOWARN)
xfs_alert(mp,
"%s : Dquot ID 0x%x (0x%p) RTBLK TIMER NOT STARTED",
str, (int)be32_to_cpu(ddq->d_id), ddq);
errs++;
}
}
}
if (!errs || !(flags & XFS_QMOPT_DQREPAIR))
return errs;
if (flags & XFS_QMOPT_DOWARN)
xfs_notice(mp, "Re-initializing dquot ID 0x%x", id);
/*
* Typically, a repair is only requested by quotacheck.
*/
ASSERT(id != -1);
ASSERT(flags & XFS_QMOPT_DQREPAIR);
memset(d, 0, sizeof(xfs_dqblk_t));
d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
d->dd_diskdq.d_flags = type;
d->dd_diskdq.d_id = cpu_to_be32(id);
if (xfs_sb_version_hascrc(&mp->m_sb)) {
uuid_copy(&d->dd_uuid, &mp->m_sb.sb_uuid);
xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
XFS_DQUOT_CRC_OFF);
}
return errs;
}
/*
* Read the disk inode attributes into the in-core inode structure.
*
* For version 5 superblocks, if we are initialising a new inode and we are not
* utilising the XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new
* inode core with a random generation number. If we are keeping inodes around,
* we need to read the inode cluster to get the existing generation number off
* disk. Further, if we are using version 4 superblocks (i.e. v1/v2 inode
* format) then log recovery is dependent on the di_flushiter field being
* initialised from the current on-disk value and hence we must also read the
* inode off disk.
*/
int
xfs_iread(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_inode_t *ip,
uint iget_flags)
{
xfs_buf_t *bp;
xfs_dinode_t *dip;
xfs_failaddr_t fa;
int error;
/*
* Fill in the location information in the in-core inode.
*/
error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
if (error)
return error;
/* shortcut IO on inode allocation if possible */
if ((iget_flags & XFS_IGET_CREATE) &&
xfs_sb_version_hascrc(&mp->m_sb) &&
!(mp->m_flags & XFS_MOUNT_IKEEP)) {
/* initialise the on-disk inode core */
memset(&ip->i_d, 0, sizeof(ip->i_d));
VFS_I(ip)->i_generation = prandom_u32();
ip->i_d.di_version = 3;
return 0;
}
/*
* Get pointers to the on-disk inode and the buffer containing it.
*/
error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &bp, 0, iget_flags);
if (error)
return error;
/* even unallocated inodes are verified */
fa = xfs_dinode_verify(mp, ip->i_ino, dip);
if (fa) {
xfs_inode_verifier_error(ip, -EFSCORRUPTED, "dinode", dip,
sizeof(*dip), fa);
error = -EFSCORRUPTED;
goto out_brelse;
}
/*
* If the on-disk inode is already linked to a directory
* entry, copy all of the inode into the in-core inode.
* xfs_iformat_fork() handles copying in the inode format
* specific information.
* Otherwise, just get the truly permanent information.
*/
if (dip->di_mode) {
xfs_inode_from_disk(ip, dip);
error = xfs_iformat_fork(ip, dip);
if (error) {
#ifdef DEBUG
xfs_alert(mp, "%s: xfs_iformat() returned error %d",
__func__, error);
#endif /* DEBUG */
goto out_brelse;
}
} else {
/*
* Partial initialisation of the in-core inode. Just the bits
* that xfs_ialloc won't overwrite or relies on being correct.
*/
ip->i_d.di_version = dip->di_version;
VFS_I(ip)->i_generation = be32_to_cpu(dip->di_gen);
ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
/*
* Make sure to pull in the mode here as well in
* case the inode is released without being used.
* This ensures that xfs_inactive() will see that
* the inode is already free and not try to mess
* with the uninitialized part of it.
*/
VFS_I(ip)->i_mode = 0;
}
ASSERT(ip->i_d.di_version >= 2);
ip->i_delayed_blks = 0;
/*
* Mark the buffer containing the inode as something to keep
* around for a while. This helps to keep recently accessed
* meta-data in-core longer.
*/
xfs_buf_set_ref(bp, XFS_INO_REF);
/*
* Use xfs_trans_brelse() to release the buffer containing the on-disk
* inode, because it was acquired with xfs_trans_read_buf() in
* xfs_imap_to_bp() above. If tp is NULL, this is just a normal
* brelse(). If we're within a transaction, then xfs_trans_brelse()
* will only release the buffer if it is not dirty within the
* transaction. It will be OK to release the buffer in this case,
* because inodes on disk are never destroyed and we will be locking the
* new in-core inode before putting it in the cache where other
* processes can find it. Thus we don't have to worry about the inode
* being changed just because we released the buffer.
*/
out_brelse:
xfs_trans_brelse(tp, bp);
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;
}
/*
* dead simple method of punching delalyed allocation blocks from a range in
* the inode. Walks a block at a time so will be slow, but is only executed in
* rare error cases so the overhead is not critical. This will always punch out
* both the start and end blocks, even if the ranges only partially overlap
* them, so it is up to the caller to ensure that partial blocks are not
* passed in.
*/
int
xfs_bmap_punch_delalloc_range(
struct xfs_inode *ip,
xfs_fileoff_t start_fsb,
xfs_fileoff_t length)
{
xfs_fileoff_t remaining = length;
int error = 0;
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
do {
int done;
xfs_bmbt_irec_t imap;
int nimaps = 1;
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
/*
* Map the range first and check that it is a delalloc extent
* before trying to unmap the range. Otherwise we will be
* trying to remove a real extent (which requires a
* transaction) or a hole, which is probably a bad idea...
*/
error = xfs_bmapi_read(ip, start_fsb, 1, &imap, &nimaps,
XFS_BMAPI_ENTIRE);
if (error) {
/* something screwed, just bail */
if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
xfs_alert(ip->i_mount,
"Failed delalloc mapping lookup ino %lld fsb %lld.",
ip->i_ino, start_fsb);
}
break;
}
if (!nimaps) {
/* nothing there */
goto next_block;
}
if (imap.br_startblock != DELAYSTARTBLOCK) {
/* been converted, ignore */
goto next_block;
}
WARN_ON(imap.br_blockcount == 0);
/*
* Note: while we initialise the firstblock/flist pair, they
* should never be used because blocks should never be
* allocated or freed for a delalloc extent and hence we need
* don't cancel or finish them after the xfs_bunmapi() call.
*/
xfs_bmap_init(&flist, &firstblock);
error = xfs_bunmapi(NULL, ip, start_fsb, 1, 0, 1, &firstblock,
&flist, &done);
if (error)
break;
ASSERT(!flist.xbf_count && !flist.xbf_first);
next_block:
start_fsb++;
remaining--;
} while(remaining > 0);
return error;
}
/*
* Read the disk inode attributes into the in-core inode structure.
*
* For version 5 superblocks, if we are initialising a new inode and we are not
* utilising the XFS_MOUNT_IKEEP inode cluster mode, we can simple build the new
* inode core with a random generation number. If we are keeping inodes around,
* we need to read the inode cluster to get the existing generation number off
* disk. Further, if we are using version 4 superblocks (i.e. v1/v2 inode
* format) then log recovery is dependent on the di_flushiter field being
* initialised from the current on-disk value and hence we must also read the
* inode off disk.
*/
int
xfs_iread(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_inode_t *ip,
uint iget_flags)
{
xfs_buf_t *bp;
xfs_dinode_t *dip;
int error;
/*
* Fill in the location information in the in-core inode.
*/
error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
if (error)
return error;
/* shortcut IO on inode allocation if possible */
if ((iget_flags & XFS_IGET_CREATE) &&
xfs_sb_version_hascrc(&mp->m_sb) &&
!(mp->m_flags & XFS_MOUNT_IKEEP)) {
/* initialise the on-disk inode core */
memset(&ip->i_d, 0, sizeof(ip->i_d));
ip->i_d.di_magic = XFS_DINODE_MAGIC;
ip->i_d.di_gen = prandom_u32();
if (xfs_sb_version_hascrc(&mp->m_sb)) {
ip->i_d.di_version = 3;
ip->i_d.di_ino = ip->i_ino;
uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_uuid);
} else
ip->i_d.di_version = 2;
return 0;
}
/*
* Get pointers to the on-disk inode and the buffer containing it.
*/
error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &dip, &bp, 0, iget_flags);
if (error)
return error;
/* even unallocated inodes are verified */
if (!xfs_dinode_verify(mp, ip, dip)) {
xfs_alert(mp, "%s: validation failed for inode %lld failed",
__func__, ip->i_ino);
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip);
error = -EFSCORRUPTED;
goto out_brelse;
}
/*
* If the on-disk inode is already linked to a directory
* entry, copy all of the inode into the in-core inode.
* xfs_iformat_fork() handles copying in the inode format
* specific information.
* Otherwise, just get the truly permanent information.
*/
if (dip->di_mode) {
xfs_dinode_from_disk(&ip->i_d, dip);
error = xfs_iformat_fork(ip, dip);
if (error) {
#ifdef DEBUG
xfs_alert(mp, "%s: xfs_iformat() returned error %d",
__func__, error);
#endif /* DEBUG */
goto out_brelse;
}
} else {
/*
* Partial initialisation of the in-core inode. Just the bits
* that xfs_ialloc won't overwrite or relies on being correct.
*/
ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
ip->i_d.di_version = dip->di_version;
ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
if (dip->di_version == 3) {
ip->i_d.di_ino = be64_to_cpu(dip->di_ino);
uuid_copy(&ip->i_d.di_uuid, &dip->di_uuid);
}
/*
* Make sure to pull in the mode here as well in
* case the inode is released without being used.
* This ensures that xfs_inactive() will see that
* the inode is already free and not try to mess
* with the uninitialized part of it.
*/
ip->i_d.di_mode = 0;
}
/*
* Automatically convert version 1 inode formats in memory to version 2
* inode format. If the inode is modified, it will get logged and
* rewritten as a version 2 inode. We can do this because we set the
* superblock feature bit for v2 inodes unconditionally during mount
* and it means the reast of the code can assume the inode version is 2
* or higher.
*/
if (ip->i_d.di_version == 1) {
ip->i_d.di_version = 2;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
ip->i_d.di_nlink = ip->i_d.di_onlink;
ip->i_d.di_onlink = 0;
xfs_set_projid(ip, 0);
}
ip->i_delayed_blks = 0;
/*
* Mark the buffer containing the inode as something to keep
* around for a while. This helps to keep recently accessed
* meta-data in-core longer.
*/
xfs_buf_set_ref(bp, XFS_INO_REF);
/*
* Use xfs_trans_brelse() to release the buffer containing the on-disk
* inode, because it was acquired with xfs_trans_read_buf() in
* xfs_imap_to_bp() above. If tp is NULL, this is just a normal
* brelse(). If we're within a transaction, then xfs_trans_brelse()
* will only release the buffer if it is not dirty within the
* transaction. It will be OK to release the buffer in this case,
* because inodes on disk are never destroyed and we will be locking the
* new in-core inode before putting it in the cache where other
* processes can find it. Thus we don't have to worry about the inode
* being changed just because we released the buffer.
*/
out_brelse:
xfs_trans_brelse(tp, bp);
return error;
}
