static int
xfs_swap_extents(
xfs_inode_t *ip, /* target inode */
xfs_inode_t *tip, /* tmp inode */
xfs_swapext_t *sxp)
{
xfs_mount_t *mp;
xfs_trans_t *tp;
xfs_bstat_t *sbp = &sxp->sx_stat;
xfs_ifork_t *tempifp, *ifp, *tifp;
int ilf_fields, tilf_fields;
int error = 0;
int aforkblks = 0;
int taforkblks = 0;
__uint64_t tmp;
mp = ip->i_mount;
tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL);
if (!tempifp) {
error = XFS_ERROR(ENOMEM);
goto out;
}
sbp = &sxp->sx_stat;
/*
* we have to do two separate lock calls here to keep lockdep
* happy. If we try to get all the locks in one call, lock will
* report false positives when we drop the ILOCK and regain them
* below.
*/
xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
/* Verify that both files have the same format */
if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
/* Verify both files are either real-time or non-realtime */
if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
if (VN_CACHED(VFS_I(tip)) != 0) {
error = xfs_flushinval_pages(tip, 0, -1,
FI_REMAPF_LOCKED);
if (error)
goto out_unlock;
}
/* Verify O_DIRECT for ftmp */
if (VN_CACHED(VFS_I(tip)) != 0) {
error = XFS_ERROR(EINVAL);
goto out_unlock;
}
/* Verify all data are being swapped */
if (sxp->sx_offset != 0 ||
sxp->sx_length != ip->i_d.di_size ||
sxp->sx_length != tip->i_d.di_size) {
error = XFS_ERROR(EFAULT);
goto out_unlock;
}
trace_xfs_swap_extent_before(ip, 0);
trace_xfs_swap_extent_before(tip, 1);
/* check inode formats now that data is flushed */
error = xfs_swap_extents_check_format(ip, tip);
if (error) {
xfs_fs_cmn_err(CE_NOTE, mp,
"%s: inode 0x%llx format is incompatible for exchanging.",
__FILE__, ip->i_ino);
goto out_unlock;
}
/*
* Compare the current change & modify times with that
* passed in. If they differ, we abort this swap.
* This is the mechanism used to ensure the calling
* process that the file was not changed out from
* under it.
*/
if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) ||
(sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
(sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
(sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
error = XFS_ERROR(EBUSY);
goto out_unlock;
}
/* We need to fail if the file is memory mapped. Once we have tossed
* all existing pages, the page fault will have no option
* but to go to the filesystem for pages. By making the page fault call
* vop_read (or write in the case of autogrow) they block on the iolock
* until we have switched the extents.
*/
if (VN_MAPPED(VFS_I(ip))) {
error = XFS_ERROR(EBUSY);
goto out_unlock;
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_iunlock(tip, XFS_ILOCK_EXCL);
/*
* There is a race condition here since we gave up the
* ilock. However, the data fork will not change since
* we have the iolock (locked for truncation too) so we
* are safe. We don't really care if non-io related
* fields change.
*/
xfs_tosspages(ip, 0, -1, FI_REMAPF);
tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
if ((error = xfs_trans_reserve(tp, 0,
XFS_ICHANGE_LOG_RES(mp), 0,
0, 0))) {
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
xfs_iunlock(tip, XFS_IOLOCK_EXCL);
xfs_trans_cancel(tp, 0);
goto out;
}
xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
/*
* Count the number of extended attribute blocks
*/
if ( ((XFS_IFORK_Q(ip) != 0) && (ip->i_d.di_anextents > 0)) &&
(ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks);
if (error)
goto out_trans_cancel;
}
if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) &&
(tip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) {
error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK,
&taforkblks);
if (error)
goto out_trans_cancel;
}
/*
* Swap the data forks of the inodes
*/
ifp = &ip->i_df;
tifp = &tip->i_df;
*tempifp = *ifp; /* struct copy */
*ifp = *tifp; /* struct copy */
*tifp = *tempifp; /* struct copy */
/*
* Fix the in-memory data fork values that are dependent on the fork
* offset in the inode. We can't assume they remain the same as attr2
* has dynamic fork offsets.
*/
ifp->if_ext_max = XFS_IFORK_SIZE(ip, XFS_DATA_FORK) /
(uint)sizeof(xfs_bmbt_rec_t);
tifp->if_ext_max = XFS_IFORK_SIZE(tip, XFS_DATA_FORK) /
(uint)sizeof(xfs_bmbt_rec_t);
/*
* Fix the on-disk inode values
*/
tmp = (__uint64_t)ip->i_d.di_nblocks;
ip->i_d.di_nblocks = tip->i_d.di_nblocks - taforkblks + aforkblks;
tip->i_d.di_nblocks = tmp + taforkblks - aforkblks;
tmp = (__uint64_t) ip->i_d.di_nextents;
ip->i_d.di_nextents = tip->i_d.di_nextents;
tip->i_d.di_nextents = tmp;
tmp = (__uint64_t) ip->i_d.di_format;
ip->i_d.di_format = tip->i_d.di_format;
tip->i_d.di_format = tmp;
/*
* The extents in the source inode could still contain speculative
* preallocation beyond EOF (e.g. the file is open but not modified
* while defrag is in progress). In that case, we need to copy over the
* number of delalloc blocks the data fork in the source inode is
* tracking beyond EOF so that when the fork is truncated away when the
* temporary inode is unlinked we don't underrun the i_delayed_blks
* counter on that inode.
*/
ASSERT(tip->i_delayed_blks == 0);
tip->i_delayed_blks = ip->i_delayed_blks;
ip->i_delayed_blks = 0;
ilf_fields = XFS_ILOG_CORE;
switch(ip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (ip->i_d.di_nextents <= XFS_INLINE_EXTS) {
ifp->if_u1.if_extents =
ifp->if_u2.if_inline_ext;
}
ilf_fields |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
ilf_fields |= XFS_ILOG_DBROOT;
break;
}
tilf_fields = XFS_ILOG_CORE;
switch(tip->i_d.di_format) {
case XFS_DINODE_FMT_EXTENTS:
/* If the extents fit in the inode, fix the
* pointer. Otherwise it's already NULL or
* pointing to the extent.
*/
if (tip->i_d.di_nextents <= XFS_INLINE_EXTS) {
tifp->if_u1.if_extents =
tifp->if_u2.if_inline_ext;
}
tilf_fields |= XFS_ILOG_DEXT;
break;
case XFS_DINODE_FMT_BTREE:
tilf_fields |= XFS_ILOG_DBROOT;
break;
}
xfs_trans_ijoin_ref(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_ijoin_ref(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_log_inode(tp, ip, ilf_fields);
xfs_trans_log_inode(tp, tip, tilf_fields);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, XFS_TRANS_SWAPEXT);
trace_xfs_swap_extent_after(ip, 0);
trace_xfs_swap_extent_after(tip, 1);
out:
kmem_free(tempifp);
return error;
out_unlock:
xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
goto out;
out_trans_cancel:
xfs_trans_cancel(tp, 0);
goto out_unlock;
}
/*
* Maps a dquot to the buffer containing its on-disk version.
* This returns a ptr to the buffer containing the on-disk dquot
* in the bpp param, and a ptr to the on-disk dquot within that buffer
*/
STATIC int
xfs_qm_dqtobp(
xfs_trans_t **tpp,
xfs_dquot_t *dqp,
xfs_disk_dquot_t **O_ddpp,
xfs_buf_t **O_bpp,
uint flags)
{
struct xfs_bmbt_irec map;
int nmaps = 1, error;
struct xfs_buf *bp;
struct xfs_inode *quotip = xfs_dq_to_quota_inode(dqp);
struct xfs_mount *mp = dqp->q_mount;
xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id);
struct xfs_trans *tp = (tpp ? *tpp : NULL);
dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
xfs_ilock(quotip, XFS_ILOCK_SHARED);
if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
/*
* Return if this type of quotas is turned off while we
* didn't have the quota inode lock.
*/
xfs_iunlock(quotip, XFS_ILOCK_SHARED);
return ESRCH;
}
/*
* Find the block map; no allocations yet
*/
error = xfs_bmapi_read(quotip, dqp->q_fileoffset,
XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0);
xfs_iunlock(quotip, XFS_ILOCK_SHARED);
if (error)
return error;
ASSERT(nmaps == 1);
ASSERT(map.br_blockcount == 1);
/*
* Offset of dquot in the (fixed sized) dquot chunk.
*/
dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
sizeof(xfs_dqblk_t);
ASSERT(map.br_startblock != DELAYSTARTBLOCK);
if (map.br_startblock == HOLESTARTBLOCK) {
/*
* We don't allocate unless we're asked to
*/
if (!(flags & XFS_QMOPT_DQALLOC))
return ENOENT;
ASSERT(tp);
error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
dqp->q_fileoffset, &bp);
if (error)
return error;
tp = *tpp;
} else {
trace_xfs_dqtobp_read(dqp);
/*
* store the blkno etc so that we don't have to do the
* mapping all the time
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0, &bp, &xfs_dquot_buf_ops);
if (error == EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
xfs_dqid_t firstid = (xfs_dqid_t)map.br_startoff *
mp->m_quotainfo->qi_dqperchunk;
ASSERT(bp == NULL);
error = xfs_qm_dqrepair(mp, tp, dqp, firstid, &bp);
}
if (error) {
ASSERT(bp == NULL);
return XFS_ERROR(error);
}
}
ASSERT(xfs_buf_islocked(bp));
*O_bpp = bp;
*O_ddpp = bp->b_addr + dqp->q_bufoffset;
return (0);
}
STATIC int
xfs_ioctl_setattr(
xfs_inode_t *ip,
struct fsxattr *fa,
int mask)
{
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
unsigned int lock_flags = 0;
struct xfs_dquot *udqp = NULL;
struct xfs_dquot *gdqp = NULL;
struct xfs_dquot *olddquot = NULL;
int code;
trace_xfs_ioctl_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
/*
* Disallow 32bit project ids when projid32bit feature is not enabled.
*/
if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) &&
!xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb))
return XFS_ERROR(EINVAL);
/*
* If disk quotas is on, we make sure that the dquots do exist on disk,
* before we start any other transactions. Trying to do this later
* is messy. We don't care to take a readlock to look at the ids
* in inode here, because we can't hold it across the trans_reserve.
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) {
code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid,
ip->i_d.di_gid, fa->fsx_projid,
XFS_QMOPT_PQUOTA, &udqp, &gdqp);
if (code)
return code;
}
/*
* For the other attributes, we acquire the inode lock and
* first do an error checking pass.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
if (code)
goto error_return;
lock_flags = XFS_ILOCK_EXCL;
xfs_ilock(ip, lock_flags);
/*
* CAP_FOWNER overrides the following restrictions:
*
* The user ID of the calling process must be equal
* to the file owner ID, except in cases where the
* CAP_FSETID capability is applicable.
*/
if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
/*
* Do a quota reservation only if projid is actually going to change.
*/
if (mask & FSX_PROJID) {
if (XFS_IS_QUOTA_RUNNING(mp) &&
XFS_IS_PQUOTA_ON(mp) &&
xfs_get_projid(ip) != fa->fsx_projid) {
ASSERT(tp);
code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (code) /* out of quota */
goto error_return;
}
}
if (mask & FSX_EXTSIZE) {
/*
* Can't change extent size if any extents are allocated.
*/
if (ip->i_d.di_nextents &&
((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
fa->fsx_extsize)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* Extent size must be a multiple of the appropriate block
* size, if set at all. It must also be smaller than the
* maximum extent size supported by the filesystem.
*
* Also, for non-realtime files, limit the extent size hint to
* half the size of the AGs in the filesystem so alignment
* doesn't result in extents larger than an AG.
*/
if (fa->fsx_extsize != 0) {
xfs_extlen_t size;
xfs_fsblock_t extsize_fsb;
extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize);
if (extsize_fsb > MAXEXTLEN) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
if (XFS_IS_REALTIME_INODE(ip) ||
((mask & FSX_XFLAGS) &&
(fa->fsx_xflags & XFS_XFLAG_REALTIME))) {
size = mp->m_sb.sb_rextsize <<
mp->m_sb.sb_blocklog;
} else {
size = mp->m_sb.sb_blocksize;
if (extsize_fsb > mp->m_sb.sb_agblocks / 2) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
if (fa->fsx_extsize % size) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
}
if (mask & FSX_XFLAGS) {
/*
* Can't change realtime flag if any extents are allocated.
*/
if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
(XFS_IS_REALTIME_INODE(ip)) !=
(fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* If realtime flag is set then must have realtime data.
*/
if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
if ((mp->m_sb.sb_rblocks == 0) ||
(mp->m_sb.sb_rextsize == 0) ||
(ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
/*
* Can't modify an immutable/append-only file unless
* we have appropriate permission.
*/
if ((ip->i_d.di_flags &
(XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) ||
(fa->fsx_xflags &
(XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
!capable(CAP_LINUX_IMMUTABLE)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
}
xfs_trans_ijoin(tp, ip, 0);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & FSX_PROJID) {
/*
* CAP_FSETID overrides the following restrictions:
*
* The set-user-ID and set-group-ID bits of a file will be
* cleared upon successful return from chown()
*/
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
/*
* Change the ownerships and register quota modifications
* in the transaction.
*/
if (xfs_get_projid(ip) != fa->fsx_projid) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) {
olddquot = xfs_qm_vop_chown(tp, ip,
&ip->i_gdquot, gdqp);
}
xfs_set_projid(ip, fa->fsx_projid);
/*
* We may have to rev the inode as well as
* the superblock version number since projids didn't
* exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
*/
if (ip->i_d.di_version == 1)
xfs_bump_ino_vers2(tp, ip);
}
}
if (mask & FSX_EXTSIZE)
ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
if (mask & FSX_XFLAGS) {
xfs_set_diflags(ip, fa->fsx_xflags);
xfs_diflags_to_linux(ip);
}
xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
* This is slightly sub-optimal in that truncates require
* two sync transactions instead of one for wsync filesystems.
* One for the truncate and one for the timestamps since we
* don't want to change the timestamps unless we're sure the
* truncate worked. Truncates are less than 1% of the laddis
* mix so this probably isn't worth the trouble to optimize.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
code = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, lock_flags);
/*
* Release any dquot(s) the inode had kept before chown.
*/
xfs_qm_dqrele(olddquot);
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
return code;
error_return:
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
xfs_trans_cancel(tp, 0);
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return code;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint resblks;
int committed;
int error;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN,
&ip->i_iocore, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
XFS_BMAP_INIT(&free_list, &firstfsb);
nimaps = 1;
error = XFS_BMAPI(mp, tp, io, offset_fsb, count_fsb,
XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
1, &imap, &nimaps, &free_list, NULL);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list), &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return XFS_ERROR(error);
if (unlikely(!imap.br_startblock &&
!(io->io_flags & XFS_IOCORE_RT)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* Handle logging requirements of various synchronous types of write.
*/
int
xfs_write_sync_logforce(
xfs_mount_t *mp,
xfs_inode_t *ip)
{
int error = 0;
/*
* If we're treating this as O_DSYNC and we have not updated the
* size, force the log.
*/
if (!(mp->m_flags & XFS_MOUNT_OSYNCISOSYNC) &&
!(ip->i_update_size)) {
xfs_inode_log_item_t *iip = ip->i_itemp;
/*
* If an allocation transaction occurred
* without extending the size, then we have to force
* the log up the proper point to ensure that the
* allocation is permanent. We can't count on
* the fact that buffered writes lock out direct I/O
* writes - the direct I/O write could have extended
* the size nontransactionally, then finished before
* we started. xfs_write_file will think that the file
* didn't grow but the update isn't safe unless the
* size change is logged.
*
* Force the log if we've committed a transaction
* against the inode or if someone else has and
* the commit record hasn't gone to disk (e.g.
* the inode is pinned). This guarantees that
* all changes affecting the inode are permanent
* when we return.
*/
if (iip && iip->ili_last_lsn) {
error = _xfs_log_force(mp, iip->ili_last_lsn,
XFS_LOG_FORCE | XFS_LOG_SYNC, NULL);
} else if (xfs_ipincount(ip) > 0) {
error = _xfs_log_force(mp, (xfs_lsn_t)0,
XFS_LOG_FORCE | XFS_LOG_SYNC, NULL);
}
} else {
xfs_trans_t *tp;
/*
* O_SYNC or O_DSYNC _with_ a size update are handled
* the same way.
*
* If the write was synchronous then we need to make
* sure that the inode modification time is permanent.
* We'll have updated the timestamp above, so here
* we use a synchronous transaction to log the inode.
* It's not fast, but it's necessary.
*
* If this a dsync write and the size got changed
* non-transactionally, then we need to ensure that
* the size change gets logged in a synchronous
* transaction.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_WRITE_SYNC);
if ((error = xfs_trans_reserve(tp, 0,
XFS_SWRITE_LOG_RES(mp),
0, 0, 0))) {
/* Transaction reserve failed */
xfs_trans_cancel(tp, 0);
} else {
/* Transaction reserve successful */
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
}
return error;
}
int
xfs_filestream_new_ag(
xfs_bmalloca_t *ap,
xfs_agnumber_t *agp)
{
int flags, err;
xfs_inode_t *ip, *pip = NULL;
xfs_mount_t *mp;
xfs_mru_cache_t *cache;
xfs_extlen_t minlen;
fstrm_item_t *dir, *file;
xfs_agnumber_t ag = NULLAGNUMBER;
ip = ap->ip;
mp = ip->i_mount;
cache = mp->m_filestream;
minlen = ap->length;
*agp = NULLAGNUMBER;
file = xfs_mru_cache_remove(cache, ip->i_ino);
if (file) {
ASSERT(ip == file->ip);
pip = file->pip;
ag = file->ag;
dir = xfs_mru_cache_lookup(cache, pip->i_ino);
if (dir) {
ASSERT(pip == dir->ip);
if (dir->ag != file->ag) {
xfs_filestream_put_ag(mp, file->ag);
xfs_filestream_get_ag(mp, dir->ag);
*agp = file->ag = dir->ag;
}
xfs_mru_cache_done(cache);
}
err = xfs_mru_cache_insert(cache, ip->i_ino, file);
if (err) {
xfs_fstrm_free_func(ip->i_ino, file);
return err;
}
if (*agp != NULLAGNUMBER) {
TRACE_MOVEAG(mp, ip, pip,
ag, xfs_filestream_peek_ag(mp, ag),
*agp, xfs_filestream_peek_ag(mp, *agp));
return 0;
}
}
if (pip)
xfs_ilock(pip, XFS_IOLOCK_EXCL | XFS_IOLOCK_PARENT);
ag = (ag == NULLAGNUMBER) ? 0 : (ag + 1) % mp->m_sb.sb_agcount;
flags = (ap->userdata ? XFS_PICK_USERDATA : 0) |
(ap->flist->xbf_low ? XFS_PICK_LOWSPACE : 0);
err = _xfs_filestream_pick_ag(mp, ag, agp, flags, minlen);
if (err || *agp == NULLAGNUMBER)
goto exit;
if (!pip) {
TRACE_ORPHAN(mp, ip, *agp);
goto exit;
}
err = _xfs_filestream_update_ag(pip, NULL, *agp);
if (err)
goto exit;
err = _xfs_filestream_update_ag(ip, pip, *agp);
if (err)
goto exit;
TRACE_MOVEAG(mp, ip, pip, NULLAGNUMBER, 0,
*agp, xfs_filestream_peek_ag(mp, *agp));
exit:
if (*agp != NULLAGNUMBER)
xfs_filestream_put_ag(mp, *agp);
else
*agp = 0;
if (pip)
xfs_iunlock(pip, XFS_IOLOCK_EXCL);
return err;
}
int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int flags,
xfs_bmbt_irec_t *ret_imap,
int *nmaps,
int found)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_filblks_t count_fsb, resaligned;
xfs_fsblock_t firstfsb;
xfs_extlen_t extsz, temp;
xfs_fsize_t isize;
int nimaps;
int bmapi_flag;
int quota_flag;
int rt;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint qblocks, resblks, resrtextents;
int committed;
int error;
/*
* Make sure that the dquots are there. This doesn't hold
* the ilock across a disk read.
*/
error = XFS_QM_DQATTACH(ip->i_mount, ip, XFS_QMOPT_ILOCKED);
if (error)
return XFS_ERROR(error);
rt = XFS_IS_REALTIME_INODE(ip);
extsz = xfs_get_extsz_hint(ip);
isize = ip->i_size;
if (io->io_new_size > isize)
isize = io->io_new_size;
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
if ((offset + count) > isize) {
error = xfs_iomap_eof_align_last_fsb(mp, io, isize, extsz,
&last_fsb);
if (error)
goto error_out;
} else {
if (found && (ret_imap->br_startblock == HOLESTARTBLOCK))
last_fsb = MIN(last_fsb, (xfs_fileoff_t)
ret_imap->br_blockcount +
ret_imap->br_startoff);
}
count_fsb = last_fsb - offset_fsb;
ASSERT(count_fsb > 0);
resaligned = count_fsb;
if (unlikely(extsz)) {
if ((temp = do_mod(offset_fsb, extsz)))
resaligned += temp;
if ((temp = do_mod(resaligned, extsz)))
resaligned += extsz - temp;
}
if (unlikely(rt)) {
resrtextents = qblocks = resaligned;
resrtextents /= mp->m_sb.sb_rextsize;
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
quota_flag = XFS_QMOPT_RES_RTBLKS;
} else {
resrtextents = 0;
resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
quota_flag = XFS_QMOPT_RES_REGBLKS;
}
/*
* Allocate and setup the transaction
*/
xfs_iunlock(ip, XFS_ILOCK_EXCL);
tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), resrtextents,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
/*
* Check for running out of space, note: need lock to return
*/
if (error)
xfs_trans_cancel(tp, 0);
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
goto error_out;
error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, tp, ip,
qblocks, 0, quota_flag);
if (error)
goto error1;
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
bmapi_flag = XFS_BMAPI_WRITE;
if ((flags & BMAPI_DIRECT) && (offset < ip->i_size || extsz))
bmapi_flag |= XFS_BMAPI_PREALLOC;
/*
* Issue the xfs_bmapi() call to allocate the blocks
*/
XFS_BMAP_INIT(&free_list, &firstfsb);
nimaps = 1;
error = XFS_BMAPI(mp, tp, io, offset_fsb, count_fsb, bmapi_flag,
&firstfsb, 0, &imap, &nimaps, &free_list, NULL);
if (error)
goto error0;
/*
* Complete the transaction
*/
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto error0;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error_out;
/*
* Copy any maps to caller's array and return any error.
*/
if (nimaps == 0) {
error = ENOSPC;
goto error_out;
}
if (unlikely(!imap.br_startblock && !(io->io_flags & XFS_IOCORE_RT))) {
error = xfs_cmn_err_fsblock_zero(ip, &imap);
goto error_out;
}
*ret_imap = imap;
*nmaps = 1;
return 0;
error0: /* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
xfs_bmap_cancel(&free_list);
XFS_TRANS_UNRESERVE_QUOTA_NBLKS(mp, tp, ip, qblocks, 0, quota_flag);
error1: /* Just cancel transaction */
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
*nmaps = 0; /* nothing set-up here */
error_out:
return XFS_ERROR(error);
}
/*
* Get a layout for the pNFS client.
*/
int
xfs_fs_map_blocks(
struct inode *inode,
loff_t offset,
u64 length,
struct iomap *iomap,
bool write,
u32 *device_generation)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
struct xfs_bmbt_irec imap;
xfs_fileoff_t offset_fsb, end_fsb;
loff_t limit;
int bmapi_flags = XFS_BMAPI_ENTIRE;
int nimaps = 1;
uint lock_flags;
int error = 0;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
/*
* We can't export inodes residing on the realtime device. The realtime
* device doesn't have a UUID to identify it, so the client has no way
* to find it.
*/
if (XFS_IS_REALTIME_INODE(ip))
return -ENXIO;
/*
* The pNFS block layout spec actually supports reflink like
* functionality, but the Linux pNFS server doesn't implement it yet.
*/
if (xfs_is_reflink_inode(ip))
return -ENXIO;
/*
* Lock out any other I/O before we flush and invalidate the pagecache,
* and then hand out a layout to the remote system. This is very
* similar to direct I/O, except that the synchronization is much more
* complicated. See the comment near xfs_break_layouts for a detailed
* explanation.
*/
xfs_ilock(ip, XFS_IOLOCK_EXCL);
error = -EINVAL;
limit = mp->m_super->s_maxbytes;
if (!write)
limit = max(limit, round_up(i_size_read(inode),
inode->i_sb->s_blocksize));
if (offset > limit)
goto out_unlock;
if (offset > limit - length)
length = limit - offset;
error = filemap_write_and_wait(inode->i_mapping);
if (error)
goto out_unlock;
error = invalidate_inode_pages2(inode->i_mapping);
if (WARN_ON_ONCE(error))
return error;
end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + length);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
lock_flags = xfs_ilock_data_map_shared(ip);
error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,
&imap, &nimaps, bmapi_flags);
xfs_iunlock(ip, lock_flags);
if (error)
goto out_unlock;
if (write) {
enum xfs_prealloc_flags flags = 0;
ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
if (!nimaps || imap.br_startblock == HOLESTARTBLOCK) {
/*
* xfs_iomap_write_direct() expects to take ownership of
* the shared ilock.
*/
xfs_ilock(ip, XFS_ILOCK_SHARED);
error = xfs_iomap_write_direct(ip, offset, length,
&imap, nimaps);
if (error)
goto out_unlock;
/*
* Ensure the next transaction is committed
* synchronously so that the blocks allocated and
* handed out to the client are guaranteed to be
* present even after a server crash.
*/
flags |= XFS_PREALLOC_SET | XFS_PREALLOC_SYNC;
}
error = xfs_update_prealloc_flags(ip, flags);
if (error)
goto out_unlock;
}
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
xfs_bmbt_to_iomap(ip, iomap, &imap);
*device_generation = mp->m_generation;
return error;
out_unlock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return error;
}
int
xfs_iomap(
xfs_inode_t *ip,
xfs_off_t offset,
ssize_t count,
int flags,
xfs_iomap_t *iomapp,
int *niomaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb, end_fsb;
int error = 0;
int lockmode = 0;
xfs_bmbt_irec_t imap;
int nimaps = 1;
int bmapi_flags = 0;
int iomap_flags = 0;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
switch (flags & (BMAPI_READ | BMAPI_WRITE | BMAPI_ALLOCATE)) {
case BMAPI_READ:
xfs_iomap_enter_trace(XFS_IOMAP_READ_ENTER, ip, offset, count);
lockmode = xfs_ilock_map_shared(ip);
bmapi_flags = XFS_BMAPI_ENTIRE;
break;
case BMAPI_WRITE:
xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, ip, offset, count);
lockmode = XFS_ILOCK_EXCL;
if (flags & BMAPI_IGNSTATE)
bmapi_flags |= XFS_BMAPI_IGSTATE|XFS_BMAPI_ENTIRE;
xfs_ilock(ip, lockmode);
break;
case BMAPI_ALLOCATE:
xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, ip, offset, count);
lockmode = XFS_ILOCK_SHARED;
bmapi_flags = XFS_BMAPI_ENTIRE;
/* Attempt non-blocking lock */
if (flags & BMAPI_TRYLOCK) {
if (!xfs_ilock_nowait(ip, lockmode))
return XFS_ERROR(EAGAIN);
} else {
xfs_ilock(ip, lockmode);
}
break;
default:
BUG();
}
ASSERT(offset <= mp->m_maxioffset);
if ((xfs_fsize_t)offset + count > mp->m_maxioffset)
count = mp->m_maxioffset - offset;
end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
error = xfs_bmapi(NULL, ip, offset_fsb,
(xfs_filblks_t)(end_fsb - offset_fsb),
bmapi_flags, NULL, 0, &imap,
&nimaps, NULL, NULL);
if (error)
goto out;
switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)) {
case BMAPI_WRITE:
/* If we found an extent, return it */
if (nimaps &&
(imap.br_startblock != HOLESTARTBLOCK) &&
(imap.br_startblock != DELAYSTARTBLOCK)) {
xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
offset, count, iomapp, &imap, flags);
break;
}
if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) {
error = xfs_iomap_write_direct(ip, offset, count, flags,
&imap, &nimaps, nimaps);
} else {
error = xfs_iomap_write_delay(ip, offset, count, flags,
&imap, &nimaps);
}
if (!error) {
xfs_iomap_map_trace(XFS_IOMAP_ALLOC_MAP, ip,
offset, count, iomapp, &imap, flags);
}
iomap_flags = IOMAP_NEW;
break;
case BMAPI_ALLOCATE:
/* If we found an extent, return it */
xfs_iunlock(ip, lockmode);
lockmode = 0;
if (nimaps && !isnullstartblock(imap.br_startblock)) {
xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
offset, count, iomapp, &imap, flags);
break;
}
error = xfs_iomap_write_allocate(ip, offset, count,
&imap, &nimaps);
break;
}
if (nimaps) {
*niomaps = xfs_imap_to_bmap(ip, offset, &imap,
iomapp, nimaps, *niomaps, iomap_flags);
} else if (niomaps) {
*niomaps = 0;
}
out:
if (lockmode)
xfs_iunlock(ip, lockmode);
return XFS_ERROR(error);
}
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 */
{
struct xfs_da_args *args;
int rval;
int v; /* type-checking value */
int lock_mode;
ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
XFS_STATS_INC(dp->i_mount, xs_dir_lookup);
/*
* We need to use KM_NOFS here so that lockdep will not throw false
* positive deadlock warnings on a non-transactional lookup path. It is
* safe to recurse into inode recalim in that case, but lockdep can't
* easily be taught about it. Hence KM_NOFS avoids having to add more
* lockdep Doing this avoids having to add a bunch of lockdep class
* annotations into the reclaim path for the ilock.
*/
args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
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->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;
lock_mode = xfs_ilock_data_map_shared(dp);
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
rval = xfs_dir2_sf_lookup(args);
goto out_check_rval;
}
rval = xfs_dir2_isblock(args, &v);
if (rval)
goto out_free;
if (v) {
rval = xfs_dir2_block_lookup(args);
goto out_check_rval;
}
rval = xfs_dir2_isleaf(args, &v);
if (rval)
goto out_free;
if (v)
rval = xfs_dir2_leaf_lookup(args);
else
rval = xfs_dir2_node_lookup(args);
out_check_rval:
if (rval == -EEXIST)
rval = 0;
if (!rval) {
*inum = args->inumber;
if (ci_name) {
ci_name->name = args->value;
ci_name->len = args->valuelen;
}
}
out_free:
xfs_iunlock(dp, lock_mode);
kmem_free(args);
return rval;
}
/*
* Make sure the blocks described by maps are stable on disk. This includes
* converting any unwritten extents, flushing the disk cache and updating the
* time stamps.
*
* Note that we rely on the caller to always send us a timestamp update so that
* we always commit a transaction here. If that stops being true we will have
* to manually flush the cache here similar to what the fsync code path does
* for datasyncs on files that have no dirty metadata.
*/
int
xfs_fs_commit_blocks(
struct inode *inode,
struct iomap *maps,
int nr_maps,
struct iattr *iattr)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
bool update_isize = false;
int error, i;
loff_t size;
ASSERT(iattr->ia_valid & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME));
xfs_ilock(ip, XFS_IOLOCK_EXCL);
size = i_size_read(inode);
if ((iattr->ia_valid & ATTR_SIZE) && iattr->ia_size > size) {
update_isize = true;
size = iattr->ia_size;
}
for (i = 0; i < nr_maps; i++) {
u64 start, length, end;
start = maps[i].offset;
if (start > size)
continue;
end = start + maps[i].length;
if (end > size)
end = size;
length = end - start;
if (!length)
continue;
/*
* Make sure reads through the pagecache see the new data.
*/
error = invalidate_inode_pages2_range(inode->i_mapping,
start >> PAGE_SHIFT,
(end - 1) >> PAGE_SHIFT);
WARN_ON_ONCE(error);
error = xfs_iomap_write_unwritten(ip, start, length);
if (error)
goto out_drop_iolock;
}
if (update_isize) {
error = xfs_pnfs_validate_isize(ip, size);
if (error)
goto out_drop_iolock;
}
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp);
if (error)
goto out_drop_iolock;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
xfs_setattr_time(ip, iattr);
if (update_isize) {
i_size_write(inode, iattr->ia_size);
ip->i_d.di_size = iattr->ia_size;
}
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp);
out_drop_iolock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return error;
}
/*
* Inodes in different states need to be treated differently, and the return
* value of xfs_iflush is not sufficient to get this right. The following table
* lists the inode states and the reclaim actions necessary for non-blocking
* reclaim:
*
*
* inode state iflush ret required action
* --------------- ---------- ---------------
* bad - reclaim
* shutdown EIO unpin and reclaim
* clean, unpinned 0 reclaim
* stale, unpinned 0 reclaim
* clean, pinned(*) 0 requeue
* stale, pinned EAGAIN requeue
* dirty, delwri ok 0 requeue
* dirty, delwri blocked EAGAIN requeue
* dirty, sync flush 0 reclaim
*
* (*) dgc: I don't think the clean, pinned state is possible but it gets
* handled anyway given the order of checks implemented.
*
* As can be seen from the table, the return value of xfs_iflush() is not
* sufficient to correctly decide the reclaim action here. The checks in
* xfs_iflush() might look like duplicates, but they are not.
*
* Also, because we get the flush lock first, we know that any inode that has
* been flushed delwri has had the flush completed by the time we check that
* the inode is clean. The clean inode check needs to be done before flushing
* the inode delwri otherwise we would loop forever requeuing clean inodes as
* we cannot tell apart a successful delwri flush and a clean inode from the
* return value of xfs_iflush().
*
* Note that because the inode is flushed delayed write by background
* writeback, the flush lock may already be held here and waiting on it can
* result in very long latencies. Hence for sync reclaims, where we wait on the
* flush lock, the caller should push out delayed write inodes first before
* trying to reclaim them to minimise the amount of time spent waiting. For
* background relaim, we just requeue the inode for the next pass.
*
* Hence the order of actions after gaining the locks should be:
* bad => reclaim
* shutdown => unpin and reclaim
* pinned, delwri => requeue
* pinned, sync => unpin
* stale => reclaim
* clean => reclaim
* dirty, delwri => flush and requeue
* dirty, sync => flush, wait and reclaim
*/
STATIC int
xfs_reclaim_inode(
struct xfs_inode *ip,
struct xfs_perag *pag,
int sync_mode)
{
int error;
restart:
error = 0;
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (!xfs_iflock_nowait(ip)) {
if (!(sync_mode & SYNC_WAIT))
goto out;
/*
* If we only have a single dirty inode in a cluster there is
* a fair chance that the AIL push may have pushed it into
* the buffer, but xfsbufd won't touch it until 30 seconds
* from now, and thus we will lock up here.
*
* Promote the inode buffer to the front of the delwri list
* and wake up xfsbufd now.
*/
xfs_promote_inode(ip);
xfs_iflock(ip);
}
if (is_bad_inode(VFS_I(ip)))
goto reclaim;
if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
xfs_iunpin_wait(ip);
goto reclaim;
}
if (xfs_ipincount(ip)) {
if (!(sync_mode & SYNC_WAIT)) {
xfs_ifunlock(ip);
goto out;
}
xfs_iunpin_wait(ip);
}
if (xfs_iflags_test(ip, XFS_ISTALE))
goto reclaim;
if (xfs_inode_clean(ip))
goto reclaim;
/*
* Now we have an inode that needs flushing.
*
* We do a nonblocking flush here even if we are doing a SYNC_WAIT
* reclaim as we can deadlock with inode cluster removal.
* xfs_ifree_cluster() can lock the inode buffer before it locks the
* ip->i_lock, and we are doing the exact opposite here. As a result,
* doing a blocking xfs_itobp() to get the cluster buffer will result
* in an ABBA deadlock with xfs_ifree_cluster().
*
* As xfs_ifree_cluser() must gather all inodes that are active in the
* cache to mark them stale, if we hit this case we don't actually want
* to do IO here - we want the inode marked stale so we can simply
* reclaim it. Hence if we get an EAGAIN error on a SYNC_WAIT flush,
* just unlock the inode, back off and try again. Hopefully the next
* pass through will see the stale flag set on the inode.
*/
error = xfs_iflush(ip, SYNC_TRYLOCK | sync_mode);
if (sync_mode & SYNC_WAIT) {
if (error == EAGAIN) {
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/* backoff longer than in xfs_ifree_cluster */
delay(2);
goto restart;
}
xfs_iflock(ip);
goto reclaim;
}
/*
* When we have to flush an inode but don't have SYNC_WAIT set, we
* flush the inode out using a delwri buffer and wait for the next
* call into reclaim to find it in a clean state instead of waiting for
* it now. We also don't return errors here - if the error is transient
* then the next reclaim pass will flush the inode, and if the error
* is permanent then the next sync reclaim will reclaim the inode and
* pass on the error.
*/
if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
xfs_warn(ip->i_mount,
"inode 0x%llx background reclaim flush failed with %d",
(long long)ip->i_ino, error);
}
out:
xfs_iflags_clear(ip, XFS_IRECLAIM);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/*
* We could return EAGAIN here to make reclaim rescan the inode tree in
* a short while. However, this just burns CPU time scanning the tree
* waiting for IO to complete and xfssyncd never goes back to the idle
* state. Instead, return 0 to let the next scheduled background reclaim
* attempt to reclaim the inode again.
*/
return 0;
reclaim:
xfs_ifunlock(ip);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
XFS_STATS_INC(xs_ig_reclaims);
/*
* Remove the inode from the per-AG radix tree.
*
* Because radix_tree_delete won't complain even if the item was never
* added to the tree assert that it's been there before to catch
* problems with the inode life time early on.
*/
spin_lock(&pag->pag_ici_lock);
if (!radix_tree_delete(&pag->pag_ici_root,
XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino)))
ASSERT(0);
__xfs_inode_clear_reclaim(pag, ip);
spin_unlock(&pag->pag_ici_lock);
/*
* Here we do an (almost) spurious inode lock in order to coordinate
* with inode cache radix tree lookups. This is because the lookup
* can reference the inodes in the cache without taking references.
*
* We make that OK here by ensuring that we wait until the inode is
* unlocked after the lookup before we go ahead and free it. We get
* both the ilock and the iolock because the code may need to drop the
* ilock one but will still hold the iolock.
*/
xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_qm_dqdetach(ip);
xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_inode_free(ip);
return error;
}
/*
* Pick a new allocation group for the current file and its file stream. This
* function is called by xfs_bmap_filestreams() with the mount point's per-ag
* lock held.
*/
int
xfs_filestream_new_ag(
xfs_bmalloca_t *ap,
xfs_agnumber_t *agp)
{
int flags, err;
xfs_inode_t *ip, *pip = NULL;
xfs_mount_t *mp;
xfs_mru_cache_t *cache;
xfs_extlen_t minlen;
fstrm_item_t *dir, *file;
xfs_agnumber_t ag = NULLAGNUMBER;
ip = ap->ip;
mp = ip->i_mount;
cache = mp->m_filestream;
minlen = ap->alen;
*agp = NULLAGNUMBER;
/*
* Look for the file in the cache, removing it if it's found. Doing
* this allows it to be held across the dir lookup that follows.
*/
file = xfs_mru_cache_remove(cache, ip->i_ino);
if (file) {
ASSERT(ip == file->ip);
/* Save the file's parent inode and old AG number for later. */
pip = file->pip;
ag = file->ag;
/* Look for the file's directory in the cache. */
dir = xfs_mru_cache_lookup(cache, pip->i_ino);
if (dir) {
ASSERT(pip == dir->ip);
/*
* If the directory has already moved on to a new AG,
* use that AG as the new AG for the file. Don't
* forget to twiddle the AG refcounts to match the
* movement.
*/
if (dir->ag != file->ag) {
xfs_filestream_put_ag(mp, file->ag);
xfs_filestream_get_ag(mp, dir->ag);
*agp = file->ag = dir->ag;
}
xfs_mru_cache_done(cache);
}
/*
* Put the file back in the cache. If this fails, the free
* function needs to be called to tidy up in the same way as if
* the item had simply expired from the cache.
*/
err = xfs_mru_cache_insert(cache, ip->i_ino, file);
if (err) {
xfs_fstrm_free_func(ip->i_ino, file);
return err;
}
/*
* If the file's AG was moved to the directory's new AG, there's
* nothing more to be done.
*/
if (*agp != NULLAGNUMBER) {
TRACE_MOVEAG(mp, ip, pip,
ag, xfs_filestream_peek_ag(mp, ag),
*agp, xfs_filestream_peek_ag(mp, *agp));
return 0;
}
}
/*
* If the file's parent directory is known, take its iolock in exclusive
* mode to prevent two sibling files from racing each other to migrate
* themselves and their parent to different AGs.
*/
if (pip)
xfs_ilock(pip, XFS_IOLOCK_EXCL);
/*
* A new AG needs to be found for the file. If the file's parent
* directory is also known, it will be moved to the new AG as well to
* ensure that files created inside it in future use the new AG.
*/
ag = (ag == NULLAGNUMBER) ? 0 : (ag + 1) % mp->m_sb.sb_agcount;
flags = (ap->userdata ? XFS_PICK_USERDATA : 0) |
(ap->low ? XFS_PICK_LOWSPACE : 0);
err = _xfs_filestream_pick_ag(mp, ag, agp, flags, minlen);
if (err || *agp == NULLAGNUMBER)
goto exit;
/*
* If the file wasn't found in the file cache, then its parent directory
* inode isn't known. For this to have happened, the file must either
* be pre-existing, or it was created long enough ago that its cache
* entry has expired. This isn't the sort of usage that the filestreams
* allocator is trying to optimise, so there's no point trying to track
* its new AG somehow in the filestream data structures.
*/
if (!pip) {
TRACE_ORPHAN(mp, ip, *agp);
goto exit;
}
/* Associate the parent inode with the AG. */
err = _xfs_filestream_update_ag(pip, NULL, *agp);
if (err)
goto exit;
/* Associate the file inode with the AG. */
err = _xfs_filestream_update_ag(ip, pip, *agp);
if (err)
goto exit;
TRACE_MOVEAG(mp, ip, pip, NULLAGNUMBER, 0,
*agp, xfs_filestream_peek_ag(mp, *agp));
exit:
/*
* If _xfs_filestream_pick_ag() returned a valid AG, remove the
* reference it took on it, since the file and directory will have taken
* their own now if they were successfully cached.
*/
if (*agp != NULLAGNUMBER)
xfs_filestream_put_ag(mp, *agp);
else
*agp = 0;
if (pip)
xfs_iunlock(pip, XFS_IOLOCK_EXCL);
return err;
}
/*
* xfs_filestream_associate() should only be called to associate a regular file
* with its parent directory. Calling it with a child directory isn't
* appropriate because filestreams don't apply to entire directory hierarchies.
* Creating a file in a child directory of an existing filestream directory
* starts a new filestream with its own allocation group association.
*
* Returns < 0 on error, 0 if successful association occurred, > 0 if
* we failed to get an association because of locking issues.
*/
int
xfs_filestream_associate(
xfs_inode_t *pip,
xfs_inode_t *ip)
{
xfs_mount_t *mp;
xfs_mru_cache_t *cache;
fstrm_item_t *item;
xfs_agnumber_t ag, rotorstep, startag;
int err = 0;
ASSERT(pip->i_d.di_mode & S_IFDIR);
ASSERT(ip->i_d.di_mode & S_IFREG);
if (!(pip->i_d.di_mode & S_IFDIR) || !(ip->i_d.di_mode & S_IFREG))
return -EINVAL;
mp = pip->i_mount;
cache = mp->m_filestream;
anon_down_read(&mp->m_peraglock);
/*
* We have a problem, Houston.
*
* Taking the iolock here violates inode locking order - we already
* hold the ilock. Hence if we block getting this lock we may never
* wake. Unfortunately, that means if we can't get the lock, we're
* screwed in terms of getting a stream association - we can't spin
* waiting for the lock because someone else is waiting on the lock we
* hold and we cannot drop that as we are in a transaction here.
*
* Lucky for us, this inversion is not a problem because it's a
* directory inode that we are trying to lock here.
*
* So, if we can't get the iolock without sleeping then just give up
*/
if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL)) {
anon_up_read(&mp->m_peraglock);
return 1;
}
/* If the parent directory is already in the cache, use its AG. */
item = xfs_mru_cache_lookup(cache, pip->i_ino);
if (item) {
ASSERT(item->ip == pip);
ag = item->ag;
xfs_mru_cache_done(cache);
TRACE_LOOKUP(mp, pip, pip, ag, xfs_filestream_peek_ag(mp, ag));
err = _xfs_filestream_update_ag(ip, pip, ag);
goto exit;
}
/*
* Set the starting AG using the rotor for inode32, otherwise
* use the directory inode's AG.
*/
if (mp->m_flags & XFS_MOUNT_32BITINODES) {
rotorstep = xfs_rotorstep;
startag = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount;
mp->m_agfrotor = (mp->m_agfrotor + 1) %
(mp->m_sb.sb_agcount * rotorstep);
} else
startag = XFS_INO_TO_AGNO(mp, pip->i_ino);
/* Pick a new AG for the parent inode starting at startag. */
err = _xfs_filestream_pick_ag(mp, startag, &ag, 0, 0);
if (err || ag == NULLAGNUMBER)
goto exit_did_pick;
/* Associate the parent inode with the AG. */
err = _xfs_filestream_update_ag(pip, NULL, ag);
if (err)
goto exit_did_pick;
/* Associate the file inode with the AG. */
err = _xfs_filestream_update_ag(ip, pip, ag);
if (err)
goto exit_did_pick;
TRACE_ASSOCIATE(mp, ip, pip, ag, xfs_filestream_peek_ag(mp, ag));
exit_did_pick:
/*
* If _xfs_filestream_pick_ag() returned a valid AG, remove the
* reference it took on it, since the file and directory will have taken
* their own now if they were successfully cached.
*/
if (ag != NULLAGNUMBER)
xfs_filestream_put_ag(mp, ag);
exit:
xfs_iunlock(pip, XFS_IOLOCK_EXCL);
anon_up_read(&mp->m_peraglock);
return -err;
}
STATIC int
xfs_ioctl_setattr(
xfs_inode_t *ip,
struct fsxattr *fa,
int mask)
{
struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
unsigned int lock_flags = 0;
struct xfs_dquot *udqp = NULL, *gdqp = NULL;
struct xfs_dquot *olddquot = NULL;
int code;
xfs_itrace_entry(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
/*
* If disk quotas is on, we make sure that the dquots do exist on disk,
* before we start any other transactions. Trying to do this later
* is messy. We don't care to take a readlock to look at the ids
* in inode here, because we can't hold it across the trans_reserve.
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) {
code = XFS_QM_DQVOPALLOC(mp, ip, ip->i_d.di_uid,
ip->i_d.di_gid, fa->fsx_projid,
XFS_QMOPT_PQUOTA, &udqp, &gdqp);
if (code)
return code;
}
/*
* For the other attributes, we acquire the inode lock and
* first do an error checking pass.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
if (code)
goto error_return;
lock_flags = XFS_ILOCK_EXCL;
xfs_ilock(ip, lock_flags);
/*
* CAP_FOWNER overrides the following restrictions:
*
* The user ID of the calling process must be equal
* to the file owner ID, except in cases where the
* CAP_FSETID capability is applicable.
*/
if (current->fsuid != ip->i_d.di_uid && !capable(CAP_FOWNER)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
/*
* Do a quota reservation only if projid is actually going to change.
*/
if (mask & FSX_PROJID) {
if (XFS_IS_PQUOTA_ON(mp) &&
ip->i_d.di_projid != fa->fsx_projid) {
ASSERT(tp);
code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (code) /* out of quota */
goto error_return;
}
}
if (mask & FSX_EXTSIZE) {
/*
* Can't change extent size if any extents are allocated.
*/
if (ip->i_d.di_nextents &&
((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
fa->fsx_extsize)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* Extent size must be a multiple of the appropriate block
* size, if set at all.
*/
if (fa->fsx_extsize != 0) {
xfs_extlen_t size;
if (XFS_IS_REALTIME_INODE(ip) ||
((mask & FSX_XFLAGS) &&
(fa->fsx_xflags & XFS_XFLAG_REALTIME))) {
size = mp->m_sb.sb_rextsize <<
mp->m_sb.sb_blocklog;
} else {
size = mp->m_sb.sb_blocksize;
}
if (fa->fsx_extsize % size) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
}
if (mask & FSX_XFLAGS) {
/*
* Can't change realtime flag if any extents are allocated.
*/
if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
(XFS_IS_REALTIME_INODE(ip)) !=
(fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
code = XFS_ERROR(EINVAL); /* EFBIG? */
goto error_return;
}
/*
* If realtime flag is set then must have realtime data.
*/
if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
if ((mp->m_sb.sb_rblocks == 0) ||
(mp->m_sb.sb_rextsize == 0) ||
(ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
code = XFS_ERROR(EINVAL);
goto error_return;
}
}
/*
* Can't modify an immutable/append-only file unless
* we have appropriate permission.
*/
if ((ip->i_d.di_flags &
(XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) ||
(fa->fsx_xflags &
(XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
!capable(CAP_LINUX_IMMUTABLE)) {
code = XFS_ERROR(EPERM);
goto error_return;
}
}
xfs_trans_ijoin(tp, ip, lock_flags);
xfs_trans_ihold(tp, ip);
/*
* Change file ownership. Must be the owner or privileged.
* If the system was configured with the "restricted_chown"
* option, the owner is not permitted to give away the file,
* and can change the group id only to a group of which he
* or she is a member.
*/
if (mask & FSX_PROJID) {
/*
* CAP_FSETID overrides the following restrictions:
*
* The set-user-ID and set-group-ID bits of a file will be
* cleared upon successful return from chown()
*/
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
/*
* Change the ownerships and register quota modifications
* in the transaction.
*/
if (ip->i_d.di_projid != fa->fsx_projid) {
if (XFS_IS_PQUOTA_ON(mp)) {
olddquot = XFS_QM_DQVOPCHOWN(mp, tp, ip,
&ip->i_gdquot, gdqp);
}
ip->i_d.di_projid = fa->fsx_projid;
/*
* We may have to rev the inode as well as
* the superblock version number since projids didn't
* exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
*/
if (ip->i_d.di_version == XFS_DINODE_VERSION_1)
xfs_bump_ino_vers2(tp, ip);
}
}
if (mask & FSX_EXTSIZE)
ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
if (mask & FSX_XFLAGS) {
xfs_set_diflags(ip, fa->fsx_xflags);
xfs_diflags_to_linux(ip);
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
XFS_STATS_INC(xs_ig_attrchg);
/*
* If this is a synchronous mount, make sure that the
* transaction goes to disk before returning to the user.
* This is slightly sub-optimal in that truncates require
* two sync transactions instead of one for wsync filesystems.
* One for the truncate and one for the timestamps since we
* don't want to change the timestamps unless we're sure the
* truncate worked. Truncates are less than 1% of the laddis
* mix so this probably isn't worth the trouble to optimize.
*/
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
code = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, lock_flags);
/*
* Release any dquot(s) the inode had kept before chown.
*/
XFS_QM_DQRELE(mp, olddquot);
XFS_QM_DQRELE(mp, udqp);
XFS_QM_DQRELE(mp, gdqp);
if (code)
return code;
if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE)) {
XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL,
NULL, DM_RIGHT_NULL, NULL, NULL, 0, 0,
(mask & FSX_NONBLOCK) ? DM_FLAGS_NDELAY : 0);
}
return 0;
error_return:
XFS_QM_DQRELE(mp, udqp);
XFS_QM_DQRELE(mp, gdqp);
xfs_trans_cancel(tp, 0);
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return code;
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*
* We no longer bother to look at the incoming map - all we have to
* guarantee is that whatever we allocate fills the required range.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap;
xfs_trans_t *tp;
int nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
xfs_bmap_init(&free_list, &first_block);
/*
* it is possible that the extents have changed since
* we did the read call as we dropped the ilock for a
* while. We have to be careful about truncates or hole
* punchs here - we are not allowed to allocate
* non-delalloc blocks here.
*
* The only protection against truncation is the pages
* for the range we are being asked to convert are
* locked and hence a truncate will block on them
* first.
*
* As a result, if we go beyond the range we really
* need and hit an delalloc extent boundary followed by
* a hole while we have excess blocks in the map, we
* will fill the hole incorrectly and overrun the
* transaction reservation.
*
* Using a single map prevents this as we are forced to
* check each map we look for overlap with the desired
* range and abort as soon as we find it. Also, given
* that we only return a single map, having one beyond
* what we can return is probably a bit silly.
*
* We also need to check that we don't go beyond EOF;
* this is a truncate optimisation as a truncate sets
* the new file size before block on the pages we
* currently have locked under writeback. Because they
* are about to be tossed, we don't need to write them
* back....
*/
nimaps = 1;
end_fsb = XFS_B_TO_FSB(mp, ip->i_size);
error = xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
if (error)
goto trans_cancel;
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = xfs_bmapi(tp, ip, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
&imap, &nimaps, &free_list, NULL);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((offset_fsb >= imap.br_startoff) &&
(offset_fsb < (imap.br_startoff +
imap.br_blockcount))) {
*map = imap;
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
/*
* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
count_fsb -= imap.br_blockcount;
map_start_fsb = imap.br_startoff + imap.br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
int
xfs_filestream_associate(
xfs_inode_t *pip,
xfs_inode_t *ip)
{
xfs_mount_t *mp;
xfs_mru_cache_t *cache;
fstrm_item_t *item;
xfs_agnumber_t ag, rotorstep, startag;
int err = 0;
ASSERT(S_ISDIR(pip->i_d.di_mode));
ASSERT(S_ISREG(ip->i_d.di_mode));
if (!S_ISDIR(pip->i_d.di_mode) || !S_ISREG(ip->i_d.di_mode))
return -EINVAL;
mp = pip->i_mount;
cache = mp->m_filestream;
if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL))
return 1;
item = xfs_mru_cache_lookup(cache, pip->i_ino);
if (item) {
ASSERT(item->ip == pip);
ag = item->ag;
xfs_mru_cache_done(cache);
TRACE_LOOKUP(mp, pip, pip, ag, xfs_filestream_peek_ag(mp, ag));
err = _xfs_filestream_update_ag(ip, pip, ag);
goto exit;
}
if (mp->m_flags & XFS_MOUNT_32BITINODES) {
rotorstep = xfs_rotorstep;
startag = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount;
mp->m_agfrotor = (mp->m_agfrotor + 1) %
(mp->m_sb.sb_agcount * rotorstep);
} else
startag = XFS_INO_TO_AGNO(mp, pip->i_ino);
err = _xfs_filestream_pick_ag(mp, startag, &ag, 0, 0);
if (err || ag == NULLAGNUMBER)
goto exit_did_pick;
err = _xfs_filestream_update_ag(pip, NULL, ag);
if (err)
goto exit_did_pick;
err = _xfs_filestream_update_ag(ip, pip, ag);
if (err)
goto exit_did_pick;
TRACE_ASSOCIATE(mp, ip, pip, ag, xfs_filestream_peek_ag(mp, ag));
exit_did_pick:
if (ag != NULLAGNUMBER)
xfs_filestream_put_ag(mp, ag);
exit:
xfs_iunlock(pip, XFS_IOLOCK_EXCL);
return -err;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
xfs_fsblock_t firstfsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
xfs_bmap_free_t free_list;
uint resblks;
int committed;
int error;
xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, ip, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
/*
* Reserve enough blocks in this transaction for two complete extent
* btree splits. We may be converting the middle part of an unwritten
* extent and in this case we will insert two new extents in the btree
* each of which could cause a full split.
*
* This reservation amount will be used in the first call to
* xfs_bmbt_split() to select an AG with enough space to satisfy the
* rest of the operation.
*/
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*/
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
error = xfs_trans_reserve(tp, resblks,
XFS_WRITE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
/*
* Modify the unwritten extent state of the buffer.
*/
xfs_bmap_init(&free_list, &firstfsb);
nimaps = 1;
error = xfs_bmapi(tp, ip, offset_fsb, count_fsb,
XFS_BMAPI_WRITE|XFS_BMAPI_CONVERT, &firstfsb,
1, &imap, &nimaps, &free_list, NULL);
if (error)
goto error_on_bmapi_transaction;
error = xfs_bmap_finish(&(tp), &(free_list), &committed);
if (error)
goto error_on_bmapi_transaction;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return XFS_ERROR(error);
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_cmn_err_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
static int
xfs_iget_cache_miss(
struct xfs_mount *mp,
struct xfs_perag *pag,
xfs_trans_t *tp,
xfs_ino_t ino,
struct xfs_inode **ipp,
int flags,
int lock_flags)
{
struct xfs_inode *ip;
int error;
xfs_agino_t agino = XFS_INO_TO_AGINO(mp, ino);
ip = xfs_inode_alloc(mp, ino);
if (!ip)
return ENOMEM;
error = xfs_iread(mp, tp, ip, flags);
if (error)
goto out_destroy;
trace_xfs_iget_miss(ip);
if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
error = ENOENT;
goto out_destroy;
}
/*
* Preload the radix tree so we can insert safely under the
* write spinlock. Note that we cannot sleep inside the preload
* region.
*/
if (radix_tree_preload(GFP_KERNEL)) {
error = EAGAIN;
goto out_destroy;
}
/*
* Because the inode hasn't been added to the radix-tree yet it can't
* be found by another thread, so we can do the non-sleeping lock here.
*/
if (lock_flags) {
if (!xfs_ilock_nowait(ip, lock_flags))
BUG();
}
/*
* These values must be set before inserting the inode into the radix
* tree as the moment it is inserted a concurrent lookup (allowed by the
* RCU locking mechanism) can find it and that lookup must see that this
* is an inode currently under construction (i.e. that XFS_INEW is set).
* The ip->i_flags_lock that protects the XFS_INEW flag forms the
* memory barrier that ensures this detection works correctly at lookup
* time.
*/
ip->i_udquot = ip->i_gdquot = NULL;
xfs_iflags_set(ip, XFS_INEW);
/* insert the new inode */
spin_lock(&pag->pag_ici_lock);
error = radix_tree_insert(&pag->pag_ici_root, agino, ip);
if (unlikely(error)) {
WARN_ON(error != -EEXIST);
XFS_STATS_INC(xs_ig_dup);
error = EAGAIN;
goto out_preload_end;
}
spin_unlock(&pag->pag_ici_lock);
radix_tree_preload_end();
*ipp = ip;
return 0;
out_preload_end:
spin_unlock(&pag->pag_ici_lock);
radix_tree_preload_end();
if (lock_flags)
xfs_iunlock(ip, lock_flags);
out_destroy:
__destroy_inode(VFS_I(ip));
xfs_inode_free(ip);
return error;
}
STATIC int
xfs_file_fsync(
struct file *file,
loff_t start,
loff_t end,
int datasync)
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
int error = 0;
int log_flushed = 0;
xfs_lsn_t lsn = 0;
trace_xfs_file_fsync(ip);
error = filemap_write_and_wait_range(inode->i_mapping, start, end);
if (error)
return error;
if (XFS_FORCED_SHUTDOWN(mp))
return -XFS_ERROR(EIO);
xfs_iflags_clear(ip, XFS_ITRUNCATED);
if (mp->m_flags & XFS_MOUNT_BARRIER) {
/*
* If we have an RT and/or log subvolume we need to make sure
* to flush the write cache the device used for file data
* first. This is to ensure newly written file data make
* it to disk before logging the new inode size in case of
* an extending write.
*/
if (XFS_IS_REALTIME_INODE(ip))
xfs_blkdev_issue_flush(mp->m_rtdev_targp);
else if (mp->m_logdev_targp != mp->m_ddev_targp)
xfs_blkdev_issue_flush(mp->m_ddev_targp);
}
/*
* All metadata updates are logged, which means that we just have
* to flush the log up to the latest LSN that touched the inode.
*/
xfs_ilock(ip, XFS_ILOCK_SHARED);
if (xfs_ipincount(ip)) {
if (!datasync ||
(ip->i_itemp->ili_fields & ~XFS_ILOG_TIMESTAMP))
lsn = ip->i_itemp->ili_last_lsn;
}
xfs_iunlock(ip, XFS_ILOCK_SHARED);
if (lsn)
error = _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, &log_flushed);
/*
* If we only have a single device, and the log force about was
* a no-op we might have to flush the data device cache here.
* This can only happen for fdatasync/O_DSYNC if we were overwriting
* an already allocated file and thus do not have any metadata to
* commit.
*/
if ((mp->m_flags & XFS_MOUNT_BARRIER) &&
mp->m_logdev_targp == mp->m_ddev_targp &&
!XFS_IS_REALTIME_INODE(ip) &&
!log_flushed)
xfs_blkdev_issue_flush(mp->m_ddev_targp);
return -error;
}
/*
* Pass in a delayed allocate extent, convert it to real extents;
* return to the caller the extent we create which maps on top of
* the originating callers request.
*
* Called without a lock on the inode.
*/
int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *map,
int *retmap)
{
xfs_mount_t *mp = ip->i_mount;
xfs_iocore_t *io = &ip->i_iocore;
xfs_fileoff_t offset_fsb, last_block;
xfs_fileoff_t end_fsb, map_start_fsb;
xfs_fsblock_t first_block;
xfs_bmap_free_t free_list;
xfs_filblks_t count_fsb;
xfs_bmbt_irec_t imap[XFS_STRAT_WRITE_IMAPS];
xfs_trans_t *tp;
int i, nimaps, committed;
int error = 0;
int nres;
*retmap = 0;
/*
* Make sure that the dquots are there.
*/
if ((error = XFS_QM_DQATTACH(mp, ip, 0)))
return XFS_ERROR(error);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = map->br_blockcount;
map_start_fsb = map->br_startoff;
XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
while (count_fsb != 0) {
/*
* Set up a transaction with which to allocate the
* backing store for the file. Do allocations in a
* loop until we get some space in the range we are
* interested in. The other space that might be allocated
* is in the delayed allocation extent on which we sit
* but before our buffer starts.
*/
nimaps = 0;
while (nimaps == 0) {
tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
tp->t_flags |= XFS_TRANS_RESERVE;
nres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK);
error = xfs_trans_reserve(tp, nres,
XFS_WRITE_LOG_RES(mp),
0, XFS_TRANS_PERM_LOG_RES,
XFS_WRITE_LOG_COUNT);
if (error) {
xfs_trans_cancel(tp, 0);
return XFS_ERROR(error);
}
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
XFS_BMAP_INIT(&free_list, &first_block);
nimaps = XFS_STRAT_WRITE_IMAPS;
/*
* Ensure we don't go beyond eof - it is possible
* the extents changed since we did the read call,
* we dropped the ilock in the interim.
*/
end_fsb = XFS_B_TO_FSB(mp, ip->i_size);
xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
last_block = XFS_FILEOFF_MAX(last_block, end_fsb);
if ((map_start_fsb + count_fsb) > last_block) {
count_fsb = last_block - map_start_fsb;
if (count_fsb == 0) {
error = EAGAIN;
goto trans_cancel;
}
}
/* Go get the actual blocks */
error = XFS_BMAPI(mp, tp, io, map_start_fsb, count_fsb,
XFS_BMAPI_WRITE, &first_block, 1,
imap, &nimaps, &free_list, NULL);
if (error)
goto trans_cancel;
error = xfs_bmap_finish(&tp, &free_list, &committed);
if (error)
goto trans_cancel;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
goto error0;
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
* See if we were able to allocate an extent that
* covers at least part of the callers request
*/
for (i = 0; i < nimaps; i++) {
if (unlikely(!imap[i].br_startblock &&
!(io->io_flags & XFS_IOCORE_RT)))
return xfs_cmn_err_fsblock_zero(ip, &imap[i]);
if ((offset_fsb >= imap[i].br_startoff) &&
(offset_fsb < (imap[i].br_startoff +
imap[i].br_blockcount))) {
*map = imap[i];
*retmap = 1;
XFS_STATS_INC(xs_xstrat_quick);
return 0;
}
count_fsb -= imap[i].br_blockcount;
}
/* So far we have not mapped the requested part of the
* file, just surrounding data, try again.
*/
nimaps--;
map_start_fsb = imap[nimaps].br_startoff +
imap[nimaps].br_blockcount;
}
trans_cancel:
xfs_bmap_cancel(&free_list);
xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
error0:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return XFS_ERROR(error);
}
/*
* Zero any on disk space between the current EOF and the new, larger EOF.
*
* This handles the normal case of zeroing the remainder of the last block in
* the file and the unusual case of zeroing blocks out beyond the size of the
* file. This second case only happens with fixed size extents and when the
* system crashes before the inode size was updated but after blocks were
* allocated.
*
* Expects the iolock to be held exclusive, and will take the ilock internally.
*/
int /* error (positive) */
xfs_zero_eof(
struct xfs_inode *ip,
xfs_off_t offset, /* starting I/O offset */
xfs_fsize_t isize) /* current inode size */
{
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t start_zero_fsb;
xfs_fileoff_t end_zero_fsb;
xfs_fileoff_t zero_count_fsb;
xfs_fileoff_t last_fsb;
xfs_fileoff_t zero_off;
xfs_fsize_t zero_len;
int nimaps;
int error = 0;
struct xfs_bmbt_irec imap;
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
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.
*/
if (XFS_B_FSB_OFFSET(mp, isize) != 0) {
error = xfs_zero_last_block(ip, offset, isize);
if (error)
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;
xfs_ilock(ip, XFS_ILOCK_EXCL);
error = xfs_bmapi_read(ip, start_zero_fsb, zero_count_fsb,
&imap, &nimaps, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
ASSERT(nimaps > 0);
if (imap.br_state == XFS_EXT_UNWRITTEN ||
imap.br_startblock == HOLESTARTBLOCK) {
start_zero_fsb = imap.br_startoff + imap.br_blockcount;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
continue;
}
/*
* There are blocks we need to zero.
*/
zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
if ((zero_off + zero_len) > offset)
zero_len = offset - zero_off;
error = xfs_iozero(ip, zero_off, zero_len);
if (error)
return error;
start_zero_fsb = imap.br_startoff + imap.br_blockcount;
ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
}
return 0;
}
STATIC int
xfs_map_blocks(
struct inode *inode,
loff_t offset,
struct xfs_bmbt_irec *imap,
int type,
int nonblocking)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
ssize_t count = 1 << inode->i_blkbits;
xfs_fileoff_t offset_fsb, end_fsb;
int error = 0;
int bmapi_flags = XFS_BMAPI_ENTIRE;
int nimaps = 1;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
if (type == XFS_IO_UNWRITTEN)
bmapi_flags |= XFS_BMAPI_IGSTATE;
if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
if (nonblocking)
return -EAGAIN;
xfs_ilock(ip, XFS_ILOCK_SHARED);
}
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
(ip->i_df.if_flags & XFS_IFEXTENTS));
ASSERT(offset <= mp->m_super->s_maxbytes);
if (offset + count > mp->m_super->s_maxbytes)
count = mp->m_super->s_maxbytes - offset;
end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb,
imap, &nimaps, bmapi_flags);
xfs_iunlock(ip, XFS_ILOCK_SHARED);
if (error)
return error;
if (type == XFS_IO_DELALLOC &&
(!nimaps || isnullstartblock(imap->br_startblock))) {
error = xfs_iomap_write_allocate(ip, offset, imap);
if (!error)
trace_xfs_map_blocks_alloc(ip, offset, count, type, imap);
return error;
}
#ifdef DEBUG
if (type == XFS_IO_UNWRITTEN) {
ASSERT(nimaps);
ASSERT(imap->br_startblock != HOLESTARTBLOCK);
ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
}
#endif
if (nimaps)
trace_xfs_map_blocks_found(ip, offset, count, type, imap);
return 0;
}
int
xfs_setattr_nonsize(
struct xfs_inode *ip,
struct iattr *iattr,
int flags)
{
xfs_mount_t *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
xfs_trans_t *tp;
int error;
kuid_t uid = GLOBAL_ROOT_UID, iuid = GLOBAL_ROOT_UID;
kgid_t gid = GLOBAL_ROOT_GID, igid = GLOBAL_ROOT_GID;
struct xfs_dquot *udqp = NULL, *gdqp = NULL;
struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL;
trace_xfs_setattr(ip);
/* If acls are being inherited, we already have this checked */
if (!(flags & XFS_ATTR_NOACL)) {
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
}
ASSERT((mask & ATTR_SIZE) == 0);
/*
* If disk quotas is on, we make sure that the dquots do exist on disk,
* before we start any other transactions. Trying to do this later
* is messy. We don't care to take a readlock to look at the ids
* in inode here, because we can't hold it across the trans_reserve.
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) {
uint qflags = 0;
if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) {
uid = iattr->ia_uid;
qflags |= XFS_QMOPT_UQUOTA;
} else {
uid = inode->i_uid;
}
if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) {
gid = iattr->ia_gid;
qflags |= XFS_QMOPT_GQUOTA;
} else {
gid = inode->i_gid;
}
/*
* We take a reference when we initialize udqp and gdqp,
* so it is important that we never blindly double trip on
* the same variable. See xfs_create() for an example.
*/
ASSERT(udqp == NULL);
ASSERT(gdqp == NULL);
error = xfs_qm_vop_dqalloc(ip, xfs_kuid_to_uid(uid),
xfs_kgid_to_gid(gid),
xfs_get_projid(ip),
qflags, &udqp, &gdqp, NULL);
if (error)
return error;
}
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
if (error)
goto out_dqrele;
xfs_ilock(ip, XFS_ILOCK_EXCL);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & (ATTR_UID|ATTR_GID)) {
/*
* These IDs could have changed since we last looked at them.
* But, we're assured that if the ownership did change
* while we didn't have the inode locked, inode's dquot(s)
* would have changed also.
*/
iuid = inode->i_uid;
igid = inode->i_gid;
gid = (mask & ATTR_GID) ? iattr->ia_gid : igid;
uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid;
/*
* Do a quota reservation only if uid/gid is actually
* going to change.
*/
if (XFS_IS_QUOTA_RUNNING(mp) &&
((XFS_IS_UQUOTA_ON(mp) && !uid_eq(iuid, uid)) ||
(XFS_IS_GQUOTA_ON(mp) && !gid_eq(igid, gid)))) {
ASSERT(tp);
error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
NULL, capable(CAP_FOWNER) ?
XFS_QMOPT_FORCE_RES : 0);
if (error) /* out of quota */
goto out_trans_cancel;
}
}
xfs_trans_ijoin(tp, ip, 0);
/*
* Change file ownership. Must be the owner or privileged.
*/
if (mask & (ATTR_UID|ATTR_GID)) {
/*
* CAP_FSETID overrides the following restrictions:
*
* The set-user-ID and set-group-ID bits of a file will be
* cleared upon successful return from chown()
*/
if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
!capable(CAP_FSETID))
ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
/*
* Change the ownerships and register quota modifications
* in the transaction.
*/
if (!uid_eq(iuid, uid)) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) {
ASSERT(mask & ATTR_UID);
ASSERT(udqp);
olddquot1 = xfs_qm_vop_chown(tp, ip,
&ip->i_udquot, udqp);
}
ip->i_d.di_uid = xfs_kuid_to_uid(uid);
inode->i_uid = uid;
}
if (!gid_eq(igid, gid)) {
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
ASSERT(!XFS_IS_PQUOTA_ON(mp));
ASSERT(mask & ATTR_GID);
ASSERT(gdqp);
olddquot2 = xfs_qm_vop_chown(tp, ip,
&ip->i_gdquot, gdqp);
}
ip->i_d.di_gid = xfs_kgid_to_gid(gid);
inode->i_gid = gid;
}
}
/*
* Change file access modes.
*/
if (mask & ATTR_MODE)
xfs_setattr_mode(tp, ip, iattr);
/*
* Change file access or modified times.
*/
if (mask & ATTR_ATIME) {
inode->i_atime = iattr->ia_atime;
ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
}
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/*
* Release any dquot(s) the inode had kept before chown.
*/
xfs_qm_dqrele(olddquot1);
xfs_qm_dqrele(olddquot2);
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
if (error)
return XFS_ERROR(error);
/*
* XXX(hch): Updating the ACL entries is not atomic vs the i_mode
* update. We could avoid this with linked transactions
* and passing down the transaction pointer all the way
* to attr_set. No previous user of the generic
* Posix ACL code seems to care about this issue either.
*/
if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
error = -xfs_acl_chmod(inode);
if (error)
return XFS_ERROR(error);
}
return 0;
out_trans_cancel:
xfs_trans_cancel(tp, 0);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
out_dqrele:
xfs_qm_dqrele(udqp);
xfs_qm_dqrele(gdqp);
return error;
}
/*
* Readdir for block directories.
*/
STATIC int
xfs_dir2_block_getdents(
struct xfs_da_args *args,
struct dir_context *ctx)
{
struct xfs_inode *dp = args->dp; /* incore directory inode */
xfs_dir2_data_hdr_t *hdr; /* block header */
struct xfs_buf *bp; /* buffer for block */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_dir2_data_unused_t *dup; /* block unused entry */
char *endptr; /* end of the data entries */
int error; /* error return value */
char *ptr; /* current data entry */
int wantoff; /* starting block offset */
xfs_off_t cook;
struct xfs_da_geometry *geo = args->geo;
int lock_mode;
/*
* If the block number in the offset is out of range, we're done.
*/
if (xfs_dir2_dataptr_to_db(geo, ctx->pos) > geo->datablk)
return 0;
lock_mode = xfs_ilock_data_map_shared(dp);
error = xfs_dir3_block_read(NULL, dp, &bp);
xfs_iunlock(dp, lock_mode);
if (error)
return error;
/*
* Extract the byte offset we start at from the seek pointer.
* We'll skip entries before this.
*/
wantoff = xfs_dir2_dataptr_to_off(geo, ctx->pos);
hdr = bp->b_addr;
xfs_dir3_data_check(dp, bp);
/*
* Set up values for the loop.
*/
btp = xfs_dir2_block_tail_p(geo, hdr);
ptr = (char *)dp->d_ops->data_entry_p(hdr);
endptr = (char *)xfs_dir2_block_leaf_p(btp);
/*
* Loop over the data portion of the block.
* Each object is a real entry (dep) or an unused one (dup).
*/
while (ptr < endptr) {
__uint8_t filetype;
dup = (xfs_dir2_data_unused_t *)ptr;
/*
* Unused, skip it.
*/
if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
ptr += be16_to_cpu(dup->length);
continue;
}
dep = (xfs_dir2_data_entry_t *)ptr;
/*
* Bump pointer for the next iteration.
*/
ptr += dp->d_ops->data_entsize(dep->namelen);
/*
* The entry is before the desired starting point, skip it.
*/
if ((char *)dep - (char *)hdr < wantoff)
continue;
cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
(char *)dep - (char *)hdr);
ctx->pos = cook & 0x7fffffff;
filetype = dp->d_ops->data_get_ftype(dep);
/*
* If it didn't fit, set the final offset to here & return.
*/
if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
be64_to_cpu(dep->inumber),
xfs_dir3_get_dtype(dp->i_mount, filetype))) {
xfs_trans_brelse(NULL, bp);
return 0;
}
}
/*
* Reached the end of the block.
* Set the offset to a non-existent block 1 and return.
*/
ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
0x7fffffff;
xfs_trans_brelse(NULL, bp);
return 0;
}
/*
* Truncate file. Must have write permission and not be a directory.
*/
int
xfs_setattr_size(
struct xfs_inode *ip,
struct iattr *iattr)
{
struct xfs_mount *mp = ip->i_mount;
struct inode *inode = VFS_I(ip);
int mask = iattr->ia_valid;
xfs_off_t oldsize, newsize;
struct xfs_trans *tp;
int error;
uint lock_flags = 0;
uint commit_flags = 0;
trace_xfs_setattr(ip);
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
error = -inode_change_ok(inode, iattr);
if (error)
return XFS_ERROR(error);
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
ASSERT(S_ISREG(ip->i_d.di_mode));
ASSERT((mask & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
oldsize = inode->i_size;
newsize = iattr->ia_size;
/*
* Short circuit the truncate case for zero length files.
*/
if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
return 0;
/*
* Use the regular setattr path to update the timestamps.
*/
iattr->ia_valid &= ~ATTR_SIZE;
return xfs_setattr_nonsize(ip, iattr, 0);
}
/*
* Make sure that the dquots are attached to the inode.
*/
error = xfs_qm_dqattach(ip, 0);
if (error)
return error;
/*
* Now we can make the changes. Before we join the inode to the
* transaction, take care of the part of the truncation that must be
* done without the inode lock. This needs to be done before joining
* the inode to the transaction, because the inode cannot be unlocked
* once it is a part of the transaction.
*/
if (newsize > oldsize) {
/*
* Do the first part of growing a file: zero any data in the
* last block that is beyond the old EOF. We need to do this
* before the inode is joined to the transaction to modify
* i_size.
*/
error = xfs_zero_eof(ip, newsize, oldsize);
if (error)
return error;
}
/*
* We are going to log the inode size change in this transaction so
* any previous writes that are beyond the on disk EOF and the new
* EOF that have not been written out need to be written here. If we
* do not write the data out, we expose ourselves to the null files
* problem.
*
* Only flush from the on disk size to the smaller of the in memory
* file size or the new size as that's the range we really care about
* here and prevents waiting for other data not within the range we
* care about here.
*/
if (oldsize != ip->i_d.di_size && newsize > ip->i_d.di_size) {
error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
ip->i_d.di_size, newsize);
if (error)
return error;
}
/*
* Wait for all direct I/O to complete.
*/
inode_dio_wait(inode);
error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
if (error)
return error;
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
if (error)
goto out_trans_cancel;
truncate_setsize(inode, newsize);
commit_flags = XFS_TRANS_RELEASE_LOG_RES;
lock_flags |= XFS_ILOCK_EXCL;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
/*
* Only change the c/mtime if we are changing the size or we are
* explicitly asked to change it. This handles the semantic difference
* between truncate() and ftruncate() as implemented in the VFS.
*
* The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
* special case where we need to update the times despite not having
* these flags set. For all other operations the VFS set these flags
* explicitly if it wants a timestamp update.
*/
if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
iattr->ia_ctime = iattr->ia_mtime =
current_fs_time(inode->i_sb);
mask |= ATTR_CTIME | ATTR_MTIME;
}
/*
* The first thing we do is set the size to new_size permanently on
* disk. This way we don't have to worry about anyone ever being able
* to look at the data being freed even in the face of a crash.
* What we're getting around here is the case where we free a block, it
* is allocated to another file, it is written to, and then we crash.
* If the new data gets written to the file but the log buffers
* containing the free and reallocation don't, then we'd end up with
* garbage in the blocks being freed. As long as we make the new size
* permanent before actually freeing any blocks it doesn't matter if
* they get written to.
*/
ip->i_d.di_size = newsize;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
if (newsize <= oldsize) {
error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, newsize);
if (error)
goto out_trans_abort;
/*
* Truncated "down", so we're removing references to old data
* here - if we delay flushing for a long time, we expose
* ourselves unduly to the notorious NULL files problem. So,
* we mark this inode and flush it when the file is closed,
* and do not wait the usual (long) time for writeout.
*/
xfs_iflags_set(ip, XFS_ITRUNCATED);
/* A truncate down always removes post-EOF blocks. */
xfs_inode_clear_eofblocks_tag(ip);
}
/*
* Change file access modes.
*/
if (mask & ATTR_MODE)
xfs_setattr_mode(tp, ip, iattr);
if (mask & ATTR_CTIME) {
inode->i_ctime = iattr->ia_ctime;
ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
}
if (mask & ATTR_MTIME) {
inode->i_mtime = iattr->ia_mtime;
ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
}
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
XFS_STATS_INC(xs_ig_attrchg);
if (mp->m_flags & XFS_MOUNT_WSYNC)
xfs_trans_set_sync(tp);
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
out_unlock:
if (lock_flags)
xfs_iunlock(ip, lock_flags);
return error;
out_trans_abort:
commit_flags |= XFS_TRANS_ABORT;
out_trans_cancel:
xfs_trans_cancel(tp, commit_flags);
goto out_unlock;
}
/*
* Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a
* a locked dquot, doing an allocation (if requested) as needed.
* When both an inode and an id are given, the inode's id takes precedence.
* That is, if the id changes while we don't hold the ilock inside this
* function, the new dquot is returned, not necessarily the one requested
* in the id argument.
*/
int
xfs_qm_dqget(
xfs_mount_t *mp,
xfs_inode_t *ip, /* locked inode (optional) */
xfs_dqid_t id, /* uid/projid/gid depending on type */
uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */
uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */
xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */
{
struct xfs_quotainfo *qi = mp->m_quotainfo;
struct radix_tree_root *tree = xfs_dquot_tree(qi, type);
struct xfs_dquot *dqp;
int error;
ASSERT(XFS_IS_QUOTA_RUNNING(mp));
if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
(! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) ||
(! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
return (ESRCH);
}
#ifdef DEBUG
if (xfs_do_dqerror) {
if ((xfs_dqerror_target == mp->m_ddev_targp) &&
(xfs_dqreq_num++ % xfs_dqerror_mod) == 0) {
xfs_debug(mp, "Returning error in dqget");
return (EIO);
}
}
ASSERT(type == XFS_DQ_USER ||
type == XFS_DQ_PROJ ||
type == XFS_DQ_GROUP);
if (ip) {
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
ASSERT(xfs_inode_dquot(ip, type) == NULL);
}
#endif
restart:
mutex_lock(&qi->qi_tree_lock);
dqp = radix_tree_lookup(tree, id);
if (dqp) {
xfs_dqlock(dqp);
if (dqp->dq_flags & XFS_DQ_FREEING) {
xfs_dqunlock(dqp);
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_freeing(dqp);
delay(1);
goto restart;
}
dqp->q_nrefs++;
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_hit(dqp);
XFS_STATS_INC(xs_qm_dqcachehits);
*O_dqpp = dqp;
return 0;
}
mutex_unlock(&qi->qi_tree_lock);
XFS_STATS_INC(xs_qm_dqcachemisses);
/*
* Dquot cache miss. We don't want to keep the inode lock across
* a (potential) disk read. Also we don't want to deal with the lock
* ordering between quotainode and this inode. OTOH, dropping the inode
* lock here means dealing with a chown that can happen before
* we re-acquire the lock.
*/
if (ip)
xfs_iunlock(ip, XFS_ILOCK_EXCL);
error = xfs_qm_dqread(mp, id, type, flags, &dqp);
if (ip)
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
if (ip) {
/*
* A dquot could be attached to this inode by now, since
* we had dropped the ilock.
*/
if (xfs_this_quota_on(mp, type)) {
struct xfs_dquot *dqp1;
dqp1 = xfs_inode_dquot(ip, type);
if (dqp1) {
xfs_qm_dqdestroy(dqp);
dqp = dqp1;
xfs_dqlock(dqp);
goto dqret;
}
} else {
/* inode stays locked on return */
xfs_qm_dqdestroy(dqp);
return XFS_ERROR(ESRCH);
}
}
mutex_lock(&qi->qi_tree_lock);
error = -radix_tree_insert(tree, id, dqp);
if (unlikely(error)) {
WARN_ON(error != EEXIST);
/*
* Duplicate found. Just throw away the new dquot and start
* over.
*/
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_dup(dqp);
xfs_qm_dqdestroy(dqp);
XFS_STATS_INC(xs_qm_dquot_dups);
goto restart;
}
/*
* We return a locked dquot to the caller, with a reference taken
*/
xfs_dqlock(dqp);
dqp->q_nrefs = 1;
qi->qi_dquots++;
mutex_unlock(&qi->qi_tree_lock);
dqret:
ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
trace_xfs_dqget_miss(dqp);
*O_dqpp = dqp;
return (0);
}
/*
* Convert userspace handle data into inode.
*
* We 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 pass that sub structure into this handy, shared routine.
*
* If no error, caller must always iput the returned inode.
*/
STATIC int
xfs_vget_fsop_handlereq(
xfs_mount_t *mp,
struct inode *parinode, /* parent inode pointer */
xfs_fsop_handlereq_t *hreq,
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;
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.fid_len !=
(hlen - sizeof(handlep->ha_fsid) -
sizeof(handlep->ha_fid.fid_len)) ||
handlep->ha_fid.fid_pad)
return XFS_ERROR(EINVAL);
}
/*
* Crack the handle, obtain the inode # & generation #
*/
xfid = (struct xfs_fid *)&handlep->ha_fid;
if (xfid->fid_len == sizeof(*xfid) - sizeof(xfid->fid_len)) {
ino = xfid->fid_ino;
igen = xfid->fid_gen;
} else {
return XFS_ERROR(EINVAL);
}
/*
* Get the XFS inode, building a Linux inode 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_gen != igen) {
xfs_iput_new(ip, XFS_ILOCK_SHARED);
return XFS_ERROR(ENOENT);
}
xfs_iunlock(ip, XFS_ILOCK_SHARED);
*inode = VFS_I(ip);
return 0;
}
/*
* Allocate a block and fill it with dquots.
* This is called when the bmapi finds a hole.
*/
STATIC int
xfs_qm_dqalloc(
xfs_trans_t **tpp,
xfs_mount_t *mp,
xfs_dquot_t *dqp,
xfs_inode_t *quotip,
xfs_fileoff_t offset_fsb,
xfs_buf_t **O_bpp)
{
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
xfs_bmbt_irec_t map;
int nmaps, error, committed;
xfs_buf_t *bp;
xfs_trans_t *tp = *tpp;
ASSERT(tp != NULL);
xfs_dqtrace_entry(dqp, "DQALLOC");
/*
* Initialize the bmap freelist prior to calling bmapi code.
*/
xfs_bmap_init(&flist, &firstblock);
xfs_ilock(quotip, XFS_ILOCK_EXCL);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (ESRCH);
}
/*
* xfs_trans_commit normally decrements the vnode ref count
* when it unlocks the inode. Since we want to keep the quota
* inode around, we bump the vnode ref count now.
*/
IHOLD(quotip);
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
if ((error = xfs_bmapi(tp, quotip,
offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB,
XFS_BMAPI_METADATA | XFS_BMAPI_WRITE,
&firstblock,
XFS_QM_DQALLOC_SPACE_RES(mp),
&map, &nmaps, &flist, NULL))) {
goto error0;
}
ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
ASSERT(nmaps == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
/*
* Keep track of the blkno to save a lookup later
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
XFS_QI_DQCHUNKLEN(mp),
0);
if (!bp || (error = XFS_BUF_GETERROR(bp)))
goto error1;
/*
* Make a chunk of dquots out of this buffer and log
* the entire thing.
*/
xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id),
dqp->dq_flags & XFS_DQ_ALLTYPES, bp);
/*
* xfs_bmap_finish() may commit the current transaction and
* start a second transaction if the freelist is not empty.
*
* Since we still want to modify this buffer, we need to
* ensure that the buffer is not released on commit of
* the first transaction and ensure the buffer is added to the
* second transaction.
*
* If there is only one transaction then don't stop the buffer
* from being released when it commits later on.
*/
xfs_trans_bhold(tp, bp);
if ((error = xfs_bmap_finish(tpp, &flist, &committed))) {
goto error1;
}
if (committed) {
tp = *tpp;
xfs_trans_bjoin(tp, bp);
} else {
xfs_trans_bhold_release(tp, bp);
}
*O_bpp = bp;
return 0;
error1:
xfs_bmap_cancel(&flist);
error0:
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return (error);
}
/*
* This gets called by xfs_trans_push_ail(), when IOP_TRYLOCK
* failed to get the inode flush lock but did get the inode locked SHARED.
* Here we're trying to see if the inode buffer is incore, and if so whether it's
* marked delayed write. If that's the case, we'll initiate a bawrite on that
* buffer to expedite the process.
*
* We aren't holding the AIL_LOCK (or the flush lock) when this gets called,
* so it is inherently race-y.
*/
STATIC void
xfs_inode_item_pushbuf(
xfs_inode_log_item_t *iip)
{
xfs_inode_t *ip;
xfs_mount_t *mp;
xfs_buf_t *bp;
uint dopush;
ip = iip->ili_inode;
ASSERT(ismrlocked(&(ip->i_lock), MR_ACCESS));
/*
* The ili_pushbuf_flag keeps others from
* trying to duplicate our effort.
*/
ASSERT(iip->ili_pushbuf_flag != 0);
ASSERT(iip->ili_push_owner == get_thread_id());
/*
* If flushlock isn't locked anymore, chances are that the
* inode flush completed and the inode was taken off the AIL.
* So, just get out.
*/
if ((valusema(&(ip->i_flock)) > 0) ||
((iip->ili_item.li_flags & XFS_LI_IN_AIL) == 0)) {
iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return;
}
mp = ip->i_mount;
bp = xfs_incore(mp->m_ddev_targ, iip->ili_format.ilf_blkno,
iip->ili_format.ilf_len, XFS_INCORE_TRYLOCK);
if (bp != NULL) {
if (XFS_BUF_ISDELAYWRITE(bp)) {
/*
* We were racing with iflush because we don't hold
* the AIL_LOCK or the flush lock. However, at this point,
* we have the buffer, and we know that it's dirty.
* So, it's possible that iflush raced with us, and
* this item is already taken off the AIL.
* If not, we can flush it async.
*/
dopush = ((iip->ili_item.li_flags & XFS_LI_IN_AIL) &&
(valusema(&(ip->i_flock)) <= 0));
iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
xfs_buftrace("INODE ITEM PUSH", bp);
if (XFS_BUF_ISPINNED(bp)) {
xfs_log_force(mp, (xfs_lsn_t)0,
XFS_LOG_FORCE);
}
if (dopush) {
xfs_bawrite(mp, bp);
} else {
xfs_buf_relse(bp);
}
} else {
iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
xfs_buf_relse(bp);
}
return;
}
/*
* We have to be careful about resetting pushbuf flag too early (above).
* Eventhough in theory we can do it as soon as we have the buflock,
* we don't want others to be doing work needlessly. They'll come to
* this function thinking that pushing the buffer is there responsibility
* only to find that the buffer is still locked by another doing the
* same thing.XXX
*/
iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return;
}
