#include "xfs_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
-#include "xfs_imap.h"
#include "xfs_trans.h"
#include "xfs_trans_priv.h"
#include "xfs_sb.h"
#include "xfs_buf_item.h"
#include "xfs_inode_item.h"
#include "xfs_btree.h"
+#include "xfs_btree_trace.h"
#include "xfs_alloc.h"
#include "xfs_ialloc.h"
#include "xfs_bmap.h"
#include "xfs_rw.h"
#include "xfs_error.h"
#include "xfs_utils.h"
-#include "xfs_dir2_trace.h"
#include "xfs_quota.h"
-#include "xfs_acl.h"
#include "xfs_filestream.h"
#include "xfs_vnodeops.h"
+#include "xfs_trace.h"
kmem_zone_t *xfs_ifork_zone;
kmem_zone_t *xfs_inode_zone;
xfs_imap_to_bp(
xfs_mount_t *mp,
xfs_trans_t *tp,
- xfs_imap_t *imap,
+ struct xfs_imap *imap,
xfs_buf_t **bpp,
uint buf_flags,
- uint imap_flags)
+ uint iget_flags)
{
int error;
int i;
"an error %d on %s. Returning error.",
error, mp->m_fsname);
} else {
- ASSERT(buf_flags & XFS_BUF_TRYLOCK);
+ ASSERT(buf_flags & XBF_TRYLOCK);
}
return error;
}
dip = (xfs_dinode_t *)xfs_buf_offset(bp,
(i << mp->m_sb.sb_inodelog));
- di_ok = be16_to_cpu(dip->di_core.di_magic) == XFS_DINODE_MAGIC &&
- XFS_DINODE_GOOD_VERSION(dip->di_core.di_version);
+ di_ok = be16_to_cpu(dip->di_magic) == XFS_DINODE_MAGIC &&
+ XFS_DINODE_GOOD_VERSION(dip->di_version);
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
- if (imap_flags & XFS_IMAP_BULKSTAT) {
+ if (iget_flags & XFS_IGET_BULKSTAT) {
xfs_trans_brelse(tp, bp);
return XFS_ERROR(EINVAL);
}
"daddr %lld #%d (magic=%x)",
XFS_BUFTARG_NAME(mp->m_ddev_targp),
(unsigned long long)imap->im_blkno, i,
- be16_to_cpu(dip->di_core.di_magic));
+ be16_to_cpu(dip->di_magic));
#endif
xfs_trans_brelse(tp, bp);
return XFS_ERROR(EFSCORRUPTED);
* Use xfs_imap() to determine the size and location of the
* buffer to read from disk.
*/
-STATIC int
+int
xfs_inotobp(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_ino_t ino,
xfs_dinode_t **dipp,
xfs_buf_t **bpp,
- int *offset)
+ int *offset,
+ uint imap_flags)
{
- xfs_imap_t imap;
+ struct xfs_imap imap;
xfs_buf_t *bp;
int error;
imap.im_blkno = 0;
- error = xfs_imap(mp, tp, ino, &imap, XFS_IMAP_LOOKUP);
+ error = xfs_imap(mp, tp, ino, &imap, imap_flags);
if (error)
return error;
- error = xfs_imap_to_bp(mp, tp, &imap, &bp, XFS_BUF_LOCK, 0);
+ error = xfs_imap_to_bp(mp, tp, &imap, &bp, XBF_LOCK, imap_flags);
if (error)
return error;
* If a non-zero error is returned, then the contents of bpp and
* dipp are undefined.
*
- * If the inode is new and has not yet been initialized, use xfs_imap()
- * to determine the size and location of the buffer to read from disk.
- * If the inode has already been mapped to its buffer and read in once,
- * then use the mapping information stored in the inode rather than
- * calling xfs_imap(). This allows us to avoid the overhead of looking
- * at the inode btree for small block file systems (see xfs_dilocate()).
- * We can tell whether the inode has been mapped in before by comparing
- * its disk block address to 0. Only uninitialized inodes will have
- * 0 for the disk block address.
+ * The inode is expected to already been mapped to its buffer and read
+ * in once, thus we can use the mapping information stored in the inode
+ * rather than calling xfs_imap(). This allows us to avoid the overhead
+ * of looking at the inode btree for small block file systems
+ * (see xfs_imap()).
*/
int
xfs_itobp(
xfs_inode_t *ip,
xfs_dinode_t **dipp,
xfs_buf_t **bpp,
- xfs_daddr_t bno,
- uint imap_flags,
uint buf_flags)
{
- xfs_imap_t imap;
xfs_buf_t *bp;
int error;
- if (ip->i_blkno == (xfs_daddr_t)0) {
- imap.im_blkno = bno;
- error = xfs_imap(mp, tp, ip->i_ino, &imap,
- XFS_IMAP_LOOKUP | imap_flags);
- if (error)
- return error;
+ ASSERT(ip->i_imap.im_blkno != 0);
- /*
- * Fill in the fields in the inode that will be used to
- * map the inode to its buffer from now on.
- */
- ip->i_blkno = imap.im_blkno;
- ip->i_len = imap.im_len;
- ip->i_boffset = imap.im_boffset;
- } else {
- /*
- * We've already mapped the inode once, so just use the
- * mapping that we saved the first time.
- */
- imap.im_blkno = ip->i_blkno;
- imap.im_len = ip->i_len;
- imap.im_boffset = ip->i_boffset;
- }
- ASSERT(bno == 0 || bno == imap.im_blkno);
-
- error = xfs_imap_to_bp(mp, tp, &imap, &bp, buf_flags, imap_flags);
+ error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp, buf_flags, 0);
if (error)
return error;
if (!bp) {
- ASSERT(buf_flags & XFS_BUF_TRYLOCK);
+ ASSERT(buf_flags & XBF_TRYLOCK);
ASSERT(tp == NULL);
*bpp = NULL;
return EAGAIN;
}
- *dipp = (xfs_dinode_t *)xfs_buf_offset(bp, imap.im_boffset);
+ *dipp = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
*bpp = bp;
return 0;
}
XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
error = 0;
- if (unlikely(be32_to_cpu(dip->di_core.di_nextents) +
- be16_to_cpu(dip->di_core.di_anextents) >
- be64_to_cpu(dip->di_core.di_nblocks))) {
+ if (unlikely(be32_to_cpu(dip->di_nextents) +
+ be16_to_cpu(dip->di_anextents) >
+ be64_to_cpu(dip->di_nblocks))) {
xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
"corrupt dinode %Lu, extent total = %d, nblocks = %Lu.",
(unsigned long long)ip->i_ino,
- (int)(be32_to_cpu(dip->di_core.di_nextents) +
- be16_to_cpu(dip->di_core.di_anextents)),
+ (int)(be32_to_cpu(dip->di_nextents) +
+ be16_to_cpu(dip->di_anextents)),
(unsigned long long)
- be64_to_cpu(dip->di_core.di_nblocks));
+ be64_to_cpu(dip->di_nblocks));
XFS_CORRUPTION_ERROR("xfs_iformat(1)", XFS_ERRLEVEL_LOW,
ip->i_mount, dip);
return XFS_ERROR(EFSCORRUPTED);
}
- if (unlikely(dip->di_core.di_forkoff > ip->i_mount->m_sb.sb_inodesize)) {
+ if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) {
xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
"corrupt dinode %Lu, forkoff = 0x%x.",
(unsigned long long)ip->i_ino,
- dip->di_core.di_forkoff);
+ dip->di_forkoff);
XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW,
ip->i_mount, dip);
return XFS_ERROR(EFSCORRUPTED);
}
+ if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
+ !ip->i_mount->m_rtdev_targp)) {
+ xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+ "corrupt dinode %Lu, has realtime flag set.",
+ ip->i_ino);
+ XFS_CORRUPTION_ERROR("xfs_iformat(realtime)",
+ XFS_ERRLEVEL_LOW, ip->i_mount, dip);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
switch (ip->i_d.di_mode & S_IFMT) {
case S_IFIFO:
case S_IFCHR:
case S_IFBLK:
case S_IFSOCK:
- if (unlikely(dip->di_core.di_format != XFS_DINODE_FMT_DEV)) {
+ if (unlikely(dip->di_format != XFS_DINODE_FMT_DEV)) {
XFS_CORRUPTION_ERROR("xfs_iformat(3)", XFS_ERRLEVEL_LOW,
ip->i_mount, dip);
return XFS_ERROR(EFSCORRUPTED);
}
ip->i_d.di_size = 0;
ip->i_size = 0;
- ip->i_df.if_u2.if_rdev = be32_to_cpu(dip->di_u.di_dev);
+ ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip);
break;
case S_IFREG:
case S_IFLNK:
case S_IFDIR:
- switch (dip->di_core.di_format) {
+ switch (dip->di_format) {
case XFS_DINODE_FMT_LOCAL:
/*
* no local regular files yet
*/
- if (unlikely((be16_to_cpu(dip->di_core.di_mode) & S_IFMT) == S_IFREG)) {
+ if (unlikely((be16_to_cpu(dip->di_mode) & S_IFMT) == S_IFREG)) {
xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
"corrupt inode %Lu "
"(local format for regular file).",
return XFS_ERROR(EFSCORRUPTED);
}
- di_size = be64_to_cpu(dip->di_core.di_size);
+ di_size = be64_to_cpu(dip->di_size);
if (unlikely(di_size > XFS_DFORK_DSIZE(dip, ip->i_mount))) {
xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
"corrupt inode %Lu "
ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
ip->i_afp->if_ext_max =
XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
- switch (dip->di_core.di_aformat) {
+ switch (dip->di_aformat) {
case XFS_DINODE_FMT_LOCAL:
atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip);
size = be16_to_cpu(atp->hdr.totsize);
+
+ if (unlikely(size < sizeof(struct xfs_attr_sf_hdr))) {
+ xfs_fs_repair_cmn_err(CE_WARN, ip->i_mount,
+ "corrupt inode %Lu "
+ "(bad attr fork size %Ld).",
+ (unsigned long long) ip->i_ino,
+ (long long) size);
+ XFS_CORRUPTION_ERROR("xfs_iformat(8)",
+ XFS_ERRLEVEL_LOW,
+ ip->i_mount, dip);
+ return XFS_ERROR(EFSCORRUPTED);
+ }
+
error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size);
break;
case XFS_DINODE_FMT_EXTENTS:
xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip));
for (i = 0; i < nex; i++, dp++) {
xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
- ep->l0 = be64_to_cpu(get_unaligned(&dp->l0));
- ep->l1 = be64_to_cpu(get_unaligned(&dp->l1));
+ ep->l0 = get_unaligned_be64(&dp->l0);
+ ep->l1 = get_unaligned_be64(&dp->l1);
}
XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork);
if (whichfork != XFS_DATA_FORK ||
ifp = XFS_IFORK_PTR(ip, whichfork);
dfp = (xfs_bmdr_block_t *)XFS_DFORK_PTR(dip, whichfork);
size = XFS_BMAP_BROOT_SPACE(dfp);
- nrecs = XFS_BMAP_BROOT_NUMRECS(dfp);
+ nrecs = be16_to_cpu(dfp->bb_numrecs);
/*
* blow out if -- fork has less extents than can fit in
* Copy and convert from the on-disk structure
* to the in-memory structure.
*/
- xfs_bmdr_to_bmbt(dfp, XFS_DFORK_SIZE(dip, ip->i_mount, whichfork),
- ifp->if_broot, size);
+ xfs_bmdr_to_bmbt(ip->i_mount, dfp,
+ XFS_DFORK_SIZE(dip, ip->i_mount, whichfork),
+ ifp->if_broot, size);
ifp->if_flags &= ~XFS_IFEXTENTS;
ifp->if_flags |= XFS_IFBROOT;
return 0;
}
-void
+STATIC void
xfs_dinode_from_disk(
xfs_icdinode_t *to,
- xfs_dinode_core_t *from)
+ xfs_dinode_t *from)
{
to->di_magic = be16_to_cpu(from->di_magic);
to->di_mode = be16_to_cpu(from->di_mode);
void
xfs_dinode_to_disk(
- xfs_dinode_core_t *to,
+ xfs_dinode_t *to,
xfs_icdinode_t *from)
{
to->di_magic = cpu_to_be16(from->di_magic);
xfs_dic2xflags(
xfs_dinode_t *dip)
{
- xfs_dinode_core_t *dic = &dip->di_core;
-
- return _xfs_dic2xflags(be16_to_cpu(dic->di_flags)) |
+ return _xfs_dic2xflags(be16_to_cpu(dip->di_flags)) |
(XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0);
}
/*
- * Given a mount structure and an inode number, return a pointer
- * to a newly allocated in-core inode corresponding to the given
- * inode number.
- *
- * Initialize the inode's attributes and extent pointers if it
- * already has them (it will not if the inode has no links).
+ * Read the disk inode attributes into the in-core inode structure.
*/
int
xfs_iread(
xfs_mount_t *mp,
xfs_trans_t *tp,
- xfs_ino_t ino,
- xfs_inode_t **ipp,
+ xfs_inode_t *ip,
xfs_daddr_t bno,
- uint imap_flags)
+ uint iget_flags)
{
xfs_buf_t *bp;
xfs_dinode_t *dip;
- xfs_inode_t *ip;
int error;
- ASSERT(xfs_inode_zone != NULL);
-
- ip = kmem_zone_zalloc(xfs_inode_zone, KM_SLEEP);
- ip->i_ino = ino;
- ip->i_mount = mp;
- atomic_set(&ip->i_iocount, 0);
- spin_lock_init(&ip->i_flags_lock);
-
/*
- * Get pointer's to the on-disk inode and the buffer containing it.
- * If the inode number refers to a block outside the file system
- * then xfs_itobp() will return NULL. In this case we should
- * return NULL as well. Set i_blkno to 0 so that xfs_itobp() will
- * know that this is a new incore inode.
+ * Fill in the location information in the in-core inode.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &bp, bno, imap_flags, XFS_BUF_LOCK);
- if (error) {
- kmem_zone_free(xfs_inode_zone, ip);
+ ip->i_imap.im_blkno = bno;
+ error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
+ if (error)
return error;
- }
+ ASSERT(bno == 0 || bno == ip->i_imap.im_blkno);
/*
- * Initialize inode's trace buffers.
- * Do this before xfs_iformat in case it adds entries.
+ * Get pointers to the on-disk inode and the buffer containing it.
*/
-#ifdef XFS_INODE_TRACE
- ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_BMAP_TRACE
- ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_BMBT_TRACE
- ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_RW_TRACE
- ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_ILOCK_TRACE
- ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_DIR2_TRACE
- ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_SLEEP);
-#endif
+ error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp,
+ XBF_LOCK, iget_flags);
+ if (error)
+ return error;
+ dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
/*
* If we got something that isn't an inode it means someone
* (nfs or dmi) has a stale handle.
*/
- if (be16_to_cpu(dip->di_core.di_magic) != XFS_DINODE_MAGIC) {
- kmem_zone_free(xfs_inode_zone, ip);
- xfs_trans_brelse(tp, bp);
+ if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC) {
#ifdef DEBUG
xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: "
- "dip->di_core.di_magic (0x%x) != "
+ "dip->di_magic (0x%x) != "
"XFS_DINODE_MAGIC (0x%x)",
- be16_to_cpu(dip->di_core.di_magic),
+ be16_to_cpu(dip->di_magic),
XFS_DINODE_MAGIC);
#endif /* DEBUG */
- return XFS_ERROR(EINVAL);
+ error = XFS_ERROR(EINVAL);
+ goto out_brelse;
}
/*
* specific information.
* Otherwise, just get the truly permanent information.
*/
- if (dip->di_core.di_mode) {
- xfs_dinode_from_disk(&ip->i_d, &dip->di_core);
+ if (dip->di_mode) {
+ xfs_dinode_from_disk(&ip->i_d, dip);
error = xfs_iformat(ip, dip);
if (error) {
- kmem_zone_free(xfs_inode_zone, ip);
- xfs_trans_brelse(tp, bp);
#ifdef DEBUG
xfs_fs_cmn_err(CE_ALERT, mp, "xfs_iread: "
"xfs_iformat() returned error %d",
error);
#endif /* DEBUG */
- return error;
+ goto out_brelse;
}
} else {
- ip->i_d.di_magic = be16_to_cpu(dip->di_core.di_magic);
- ip->i_d.di_version = dip->di_core.di_version;
- ip->i_d.di_gen = be32_to_cpu(dip->di_core.di_gen);
- ip->i_d.di_flushiter = be16_to_cpu(dip->di_core.di_flushiter);
+ ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
+ ip->i_d.di_version = dip->di_version;
+ ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
+ ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
/*
* Make sure to pull in the mode here as well in
* case the inode is released without being used.
XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
}
- INIT_LIST_HEAD(&ip->i_reclaim);
-
/*
* The inode format changed when we moved the link count and
* made it 32 bits long. If this is an old format inode,
* the new format. We don't change the version number so that we
* can distinguish this from a real new format inode.
*/
- if (ip->i_d.di_version == XFS_DINODE_VERSION_1) {
+ if (ip->i_d.di_version == 1) {
ip->i_d.di_nlink = ip->i_d.di_onlink;
ip->i_d.di_onlink = 0;
ip->i_d.di_projid = 0;
* around for a while. This helps to keep recently accessed
* meta-data in-core longer.
*/
- XFS_BUF_SET_REF(bp, XFS_INO_REF);
+ XFS_BUF_SET_REF(bp, XFS_INO_REF);
/*
* Use xfs_trans_brelse() to release the buffer containing the
* to worry about the inode being changed just because we released
* the buffer.
*/
+ out_brelse:
xfs_trans_brelse(tp, bp);
- *ipp = ip;
- return 0;
+ return error;
}
/*
{
xfs_ino_t ino;
xfs_inode_t *ip;
- bhv_vnode_t *vp;
uint flags;
int error;
+ timespec_t tv;
+ int filestreams = 0;
/*
* Call the space management code to pick
*/
error = xfs_dialloc(tp, pip ? pip->i_ino : 0, mode, okalloc,
ialloc_context, call_again, &ino);
- if (error != 0) {
+ if (error)
return error;
- }
if (*call_again || ino == NULLFSINO) {
*ipp = NULL;
return 0;
*/
error = xfs_trans_iget(tp->t_mountp, tp, ino,
XFS_IGET_CREATE, XFS_ILOCK_EXCL, &ip);
- if (error != 0) {
+ if (error)
return error;
- }
ASSERT(ip != NULL);
- vp = VFS_I(ip);
ip->i_d.di_mode = (__uint16_t)mode;
ip->i_d.di_onlink = 0;
ip->i_d.di_nlink = nlink;
ASSERT(ip->i_d.di_nlink == nlink);
- ip->i_d.di_uid = current_fsuid(cr);
- ip->i_d.di_gid = current_fsgid(cr);
+ ip->i_d.di_uid = current_fsuid();
+ ip->i_d.di_gid = current_fsgid();
ip->i_d.di_projid = prid;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
* here rather than here and in the flush/logging code.
*/
if (xfs_sb_version_hasnlink(&tp->t_mountp->m_sb) &&
- ip->i_d.di_version == XFS_DINODE_VERSION_1) {
- ip->i_d.di_version = XFS_DINODE_VERSION_2;
+ ip->i_d.di_version == 1) {
+ ip->i_d.di_version = 2;
/*
* We've already zeroed the old link count, the projid field,
* and the pad field.
/*
* Project ids won't be stored on disk if we are using a version 1 inode.
*/
- if ((prid != 0) && (ip->i_d.di_version == XFS_DINODE_VERSION_1))
+ if ((prid != 0) && (ip->i_d.di_version == 1))
xfs_bump_ino_vers2(tp, ip);
if (pip && XFS_INHERIT_GID(pip)) {
ip->i_size = 0;
ip->i_d.di_nextents = 0;
ASSERT(ip->i_d.di_nblocks == 0);
- xfs_ichgtime(ip, XFS_ICHGTIME_CHG|XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD);
+
+ nanotime(&tv);
+ ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
+ ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
+ ip->i_d.di_atime = ip->i_d.di_mtime;
+ ip->i_d.di_ctime = ip->i_d.di_mtime;
+
/*
* di_gen will have been taken care of in xfs_iread.
*/
flags |= XFS_ILOG_DEV;
break;
case S_IFREG:
- if (pip && xfs_inode_is_filestream(pip)) {
- error = xfs_filestream_associate(pip, ip);
- if (error < 0)
- return -error;
- if (!error)
- xfs_iflags_set(ip, XFS_IFILESTREAM);
- }
+ /*
+ * we can't set up filestreams until after the VFS inode
+ * is set up properly.
+ */
+ if (pip && xfs_inode_is_filestream(pip))
+ filestreams = 1;
/* fall through */
case S_IFDIR:
if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
xfs_trans_log_inode(tp, ip, flags);
/* now that we have an i_mode we can setup inode ops and unlock */
- xfs_initialize_vnode(tp->t_mountp, vp, ip);
+ xfs_setup_inode(ip);
+
+ /* now we have set up the vfs inode we can associate the filestream */
+ if (filestreams) {
+ error = xfs_filestream_associate(pip, ip);
+ if (error < 0)
+ return -error;
+ if (!error)
+ xfs_iflags_set(ip, XFS_IFILESTREAM);
+ }
*ipp = ip;
return 0;
* In that case the pages will still be in memory, but the inode size
* will never have been updated.
*/
-xfs_fsize_t
+STATIC xfs_fsize_t
xfs_file_last_byte(
xfs_inode_t *ip)
{
* necessary.
*/
if (ip->i_df.if_flags & XFS_IFEXTENTS) {
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
error = xfs_bmap_last_offset(NULL, ip, &last_block,
XFS_DATA_FORK);
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
if (error) {
last_block = 0;
}
return last_byte;
}
-#if defined(XFS_RW_TRACE)
-STATIC void
-xfs_itrunc_trace(
- int tag,
- xfs_inode_t *ip,
- int flag,
- xfs_fsize_t new_size,
- xfs_off_t toss_start,
- xfs_off_t toss_finish)
-{
- if (ip->i_rwtrace == NULL) {
- return;
- }
-
- ktrace_enter(ip->i_rwtrace,
- (void*)((long)tag),
- (void*)ip,
- (void*)(unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff),
- (void*)(unsigned long)(ip->i_d.di_size & 0xffffffff),
- (void*)((long)flag),
- (void*)(unsigned long)((new_size >> 32) & 0xffffffff),
- (void*)(unsigned long)(new_size & 0xffffffff),
- (void*)(unsigned long)((toss_start >> 32) & 0xffffffff),
- (void*)(unsigned long)(toss_start & 0xffffffff),
- (void*)(unsigned long)((toss_finish >> 32) & 0xffffffff),
- (void*)(unsigned long)(toss_finish & 0xffffffff),
- (void*)(unsigned long)current_cpu(),
- (void*)(unsigned long)current_pid(),
- (void*)NULL,
- (void*)NULL,
- (void*)NULL);
-}
-#else
-#define xfs_itrunc_trace(tag, ip, flag, new_size, toss_start, toss_finish)
-#endif
-
/*
* Start the truncation of the file to new_size. The new size
* must be smaller than the current size. This routine will
* direct I/O with the truncate operation. Also, because we hold
* the IOLOCK in exclusive mode, we prevent new direct I/Os from being
* started until the truncate completes and drops the lock. Essentially,
- * the vn_iowait() call forms an I/O barrier that provides strict ordering
- * between direct I/Os and the truncate operation.
+ * the xfs_ioend_wait() call forms an I/O barrier that provides strict
+ * ordering between direct I/Os and the truncate operation.
*
* The flags parameter can have either the value XFS_ITRUNC_DEFINITE
* or XFS_ITRUNC_MAYBE. The XFS_ITRUNC_MAYBE value should be used
mp = ip->i_mount;
/* wait for the completion of any pending DIOs */
- if (new_size < ip->i_size)
- vn_iowait(ip);
+ if (new_size == 0 || new_size < ip->i_size)
+ xfs_ioend_wait(ip);
/*
* Call toss_pages or flushinval_pages to get rid of pages
return 0;
}
last_byte = xfs_file_last_byte(ip);
- xfs_itrunc_trace(XFS_ITRUNC_START, ip, flags, new_size, toss_start,
- last_byte);
+ trace_xfs_itruncate_start(ip, flags, new_size, toss_start, last_byte);
if (last_byte > toss_start) {
if (flags & XFS_ITRUNC_DEFINITE) {
xfs_tosspages(ip, toss_start,
new_size = 0LL;
}
first_unmap_block = XFS_B_TO_FSB(mp, (xfs_ufsize_t)new_size);
- xfs_itrunc_trace(XFS_ITRUNC_FINISH1, ip, 0, new_size, 0, 0);
+ trace_xfs_itruncate_finish_start(ip, new_size);
+
/*
* 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
* in this file with garbage in them once recovery
* runs.
*/
- XFS_BMAP_INIT(&free_list, &first_block);
+ xfs_bmap_init(&free_list, &first_block);
error = xfs_bunmapi(ntp, ip,
first_unmap_block, unmap_len,
- XFS_BMAPI_AFLAG(fork) |
+ xfs_bmapi_aflag(fork) |
(sync ? 0 : XFS_BMAPI_ASYNC),
XFS_ITRUNC_MAX_EXTENTS,
&first_block, &free_list,
xfs_trans_ijoin(ntp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
xfs_trans_ihold(ntp, ip);
- if (!error)
- error = xfs_trans_reserve(ntp, 0,
+ if (error)
+ return error;
+ /*
+ * transaction commit worked ok so we can drop the extra ticket
+ * reference that we gained in xfs_trans_dup()
+ */
+ xfs_log_ticket_put(ntp->t_ticket);
+ error = xfs_trans_reserve(ntp, 0,
XFS_ITRUNCATE_LOG_RES(mp), 0,
XFS_TRANS_PERM_LOG_RES,
XFS_ITRUNCATE_LOG_COUNT);
ASSERT((new_size != 0) ||
(fork == XFS_ATTR_FORK) ||
(ip->i_d.di_nextents == 0));
- xfs_itrunc_trace(XFS_ITRUNC_FINISH2, ip, 0, new_size, 0, 0);
+ trace_xfs_itruncate_finish_end(ip, new_size);
return 0;
}
xfs_dinode_t *dip;
xfs_buf_t *agibp;
xfs_buf_t *ibp;
- xfs_agnumber_t agno;
- xfs_daddr_t agdaddr;
xfs_agino_t agino;
short bucket_index;
int offset;
int error;
- int agi_ok;
ASSERT(ip->i_d.di_nlink == 0);
ASSERT(ip->i_d.di_mode != 0);
mp = tp->t_mountp;
- agno = XFS_INO_TO_AGNO(mp, ip->i_ino);
- agdaddr = XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp));
-
/*
* Get the agi buffer first. It ensures lock ordering
* on the list.
*/
- error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, agdaddr,
- XFS_FSS_TO_BB(mp, 1), 0, &agibp);
+ error = xfs_read_agi(mp, tp, XFS_INO_TO_AGNO(mp, ip->i_ino), &agibp);
if (error)
return error;
-
- /*
- * Validate the magic number of the agi block.
- */
agi = XFS_BUF_TO_AGI(agibp);
- agi_ok =
- be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC &&
- XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
- if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK,
- XFS_RANDOM_IUNLINK))) {
- XFS_CORRUPTION_ERROR("xfs_iunlink", XFS_ERRLEVEL_LOW, mp, agi);
- xfs_trans_brelse(tp, agibp);
- return XFS_ERROR(EFSCORRUPTED);
- }
+
/*
* Get the index into the agi hash table for the
* list this inode will go on.
* Here we put the head pointer into our next pointer,
* and then we fall through to point the head at us.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error)
return error;
ASSERT(be32_to_cpu(dip->di_next_unlinked) == NULLAGINO);
/* both on-disk, don't endian flip twice */
dip->di_next_unlinked = agi->agi_unlinked[bucket_index];
- offset = ip->i_boffset +
+ offset = ip->i_imap.im_boffset +
offsetof(xfs_dinode_t, di_next_unlinked);
xfs_trans_inode_buf(tp, ibp);
xfs_trans_log_buf(tp, ibp, offset,
xfs_buf_t *agibp;
xfs_buf_t *ibp;
xfs_agnumber_t agno;
- xfs_daddr_t agdaddr;
xfs_agino_t agino;
xfs_agino_t next_agino;
xfs_buf_t *last_ibp;
short bucket_index;
int offset, last_offset = 0;
int error;
- int agi_ok;
- /*
- * First pull the on-disk inode from the AGI unlinked list.
- */
mp = tp->t_mountp;
-
agno = XFS_INO_TO_AGNO(mp, ip->i_ino);
- agdaddr = XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp));
/*
* Get the agi buffer first. It ensures lock ordering
* on the list.
*/
- error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, agdaddr,
- XFS_FSS_TO_BB(mp, 1), 0, &agibp);
- if (error) {
- cmn_err(CE_WARN,
- "xfs_iunlink_remove: xfs_trans_read_buf() returned an error %d on %s. Returning error.",
- error, mp->m_fsname);
+ error = xfs_read_agi(mp, tp, agno, &agibp);
+ if (error)
return error;
- }
- /*
- * Validate the magic number of the agi block.
- */
+
agi = XFS_BUF_TO_AGI(agibp);
- agi_ok =
- be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC &&
- XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum));
- if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK_REMOVE,
- XFS_RANDOM_IUNLINK_REMOVE))) {
- XFS_CORRUPTION_ERROR("xfs_iunlink_remove", XFS_ERRLEVEL_LOW,
- mp, agi);
- xfs_trans_brelse(tp, agibp);
- cmn_err(CE_WARN,
- "xfs_iunlink_remove: XFS_TEST_ERROR() returned an error on %s. Returning EFSCORRUPTED.",
- mp->m_fsname);
- return XFS_ERROR(EFSCORRUPTED);
- }
+
/*
* Get the index into the agi hash table for the
* list this inode will go on.
* of dealing with the buffer when there is no need to
* change it.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
ASSERT(next_agino != 0);
if (next_agino != NULLAGINO) {
dip->di_next_unlinked = cpu_to_be32(NULLAGINO);
- offset = ip->i_boffset +
+ offset = ip->i_imap.im_boffset +
offsetof(xfs_dinode_t, di_next_unlinked);
xfs_trans_inode_buf(tp, ibp);
xfs_trans_log_buf(tp, ibp, offset,
}
next_ino = XFS_AGINO_TO_INO(mp, agno, next_agino);
error = xfs_inotobp(mp, tp, next_ino, &last_dip,
- &last_ibp, &last_offset);
+ &last_ibp, &last_offset, 0);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_inotobp() returned an error %d on %s. Returning error.",
* Now last_ibp points to the buffer previous to us on
* the unlinked list. Pull us from the list.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
ASSERT(next_agino != agino);
if (next_agino != NULLAGINO) {
dip->di_next_unlinked = cpu_to_be32(NULLAGINO);
- offset = ip->i_boffset +
+ offset = ip->i_imap.im_boffset +
offsetof(xfs_dinode_t, di_next_unlinked);
xfs_trans_inode_buf(tp, ibp);
xfs_trans_log_buf(tp, ibp, offset,
xfs_inode_t *ip, **ip_found;
xfs_inode_log_item_t *iip;
xfs_log_item_t *lip;
- xfs_perag_t *pag = xfs_get_perag(mp, inum);
+ struct xfs_perag *pag;
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inum));
if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
blks_per_cluster = 1;
ninodes = mp->m_sb.sb_inopblock;
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
mp->m_bsize * blks_per_cluster,
- XFS_BUF_LOCK);
+ XBF_LOCK);
pre_flushed = 0;
lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
iip = (xfs_inode_log_item_t *)lip;
ASSERT(iip->ili_logged == 1);
lip->li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*)) xfs_istale_done;
- spin_lock(&mp->m_ail_lock);
- iip->ili_flush_lsn = iip->ili_item.li_lsn;
- spin_unlock(&mp->m_ail_lock);
+ xfs_trans_ail_copy_lsn(mp->m_ail,
+ &iip->ili_flush_lsn,
+ &iip->ili_item.li_lsn);
xfs_iflags_set(iip->ili_inode, XFS_ISTALE);
pre_flushed++;
}
iip->ili_last_fields = iip->ili_format.ilf_fields;
iip->ili_format.ilf_fields = 0;
iip->ili_logged = 1;
- spin_lock(&mp->m_ail_lock);
- iip->ili_flush_lsn = iip->ili_item.li_lsn;
- spin_unlock(&mp->m_ail_lock);
+ xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
+ &iip->ili_item.li_lsn);
xfs_buf_attach_iodone(bp,
(void(*)(xfs_buf_t*,xfs_log_item_t*))
}
kmem_free(ip_found);
- xfs_put_perag(mp, pag);
+ xfs_perag_put(pag);
}
/*
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, 0, 0, XFS_BUF_LOCK);
+ error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, XBF_LOCK);
if (error)
return error;
* This is a temporary hack that would require a proper fix
* in the future.
*/
- dip->di_core.di_mode = 0;
+ dip->di_mode = 0;
if (delete) {
xfs_ifree_cluster(ip, tp, first_ino);
int rec_diff,
int whichfork)
{
+ struct xfs_mount *mp = ip->i_mount;
int cur_max;
xfs_ifork_t *ifp;
- xfs_bmbt_block_t *new_broot;
+ struct xfs_btree_block *new_broot;
int new_max;
size_t new_size;
char *np;
*/
if (ifp->if_broot_bytes == 0) {
new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(rec_diff);
- ifp->if_broot = (xfs_bmbt_block_t*)kmem_alloc(new_size,
- KM_SLEEP);
+ ifp->if_broot = kmem_alloc(new_size, KM_SLEEP);
ifp->if_broot_bytes = (int)new_size;
return;
}
* location. The records don't change location because
* they are kept butted up against the btree block header.
*/
- cur_max = XFS_BMAP_BROOT_MAXRECS(ifp->if_broot_bytes);
+ cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0);
new_max = cur_max + rec_diff;
new_size = (size_t)XFS_BMAP_BROOT_SPACE_CALC(new_max);
- ifp->if_broot = (xfs_bmbt_block_t *)
- kmem_realloc(ifp->if_broot,
- new_size,
+ ifp->if_broot = kmem_realloc(ifp->if_broot, new_size,
(size_t)XFS_BMAP_BROOT_SPACE_CALC(cur_max), /* old size */
KM_SLEEP);
- op = (char *)XFS_BMAP_BROOT_PTR_ADDR(ifp->if_broot, 1,
- ifp->if_broot_bytes);
- np = (char *)XFS_BMAP_BROOT_PTR_ADDR(ifp->if_broot, 1,
- (int)new_size);
+ op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
+ ifp->if_broot_bytes);
+ np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
+ (int)new_size);
ifp->if_broot_bytes = (int)new_size;
ASSERT(ifp->if_broot_bytes <=
XFS_IFORK_SIZE(ip, whichfork) + XFS_BROOT_SIZE_ADJ);
* records, just get rid of the root and clear the status bit.
*/
ASSERT((ifp->if_broot != NULL) && (ifp->if_broot_bytes > 0));
- cur_max = XFS_BMAP_BROOT_MAXRECS(ifp->if_broot_bytes);
+ cur_max = xfs_bmbt_maxrecs(mp, ifp->if_broot_bytes, 0);
new_max = cur_max + rec_diff;
ASSERT(new_max >= 0);
if (new_max > 0)
else
new_size = 0;
if (new_size > 0) {
- new_broot = (xfs_bmbt_block_t *)kmem_alloc(new_size, KM_SLEEP);
+ new_broot = kmem_alloc(new_size, KM_SLEEP);
/*
* First copy over the btree block header.
*/
- memcpy(new_broot, ifp->if_broot, sizeof(xfs_bmbt_block_t));
+ memcpy(new_broot, ifp->if_broot, XFS_BTREE_LBLOCK_LEN);
} else {
new_broot = NULL;
ifp->if_flags &= ~XFS_IFBROOT;
/*
* First copy the records.
*/
- op = (char *)XFS_BMAP_BROOT_REC_ADDR(ifp->if_broot, 1,
- ifp->if_broot_bytes);
- np = (char *)XFS_BMAP_BROOT_REC_ADDR(new_broot, 1,
- (int)new_size);
+ op = (char *)XFS_BMBT_REC_ADDR(mp, ifp->if_broot, 1);
+ np = (char *)XFS_BMBT_REC_ADDR(mp, new_broot, 1);
memcpy(np, op, new_max * (uint)sizeof(xfs_bmbt_rec_t));
/*
* Then copy the pointers.
*/
- op = (char *)XFS_BMAP_BROOT_PTR_ADDR(ifp->if_broot, 1,
+ op = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, ifp->if_broot, 1,
ifp->if_broot_bytes);
- np = (char *)XFS_BMAP_BROOT_PTR_ADDR(new_broot, 1,
+ np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, new_broot, 1,
(int)new_size);
memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t));
}
ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork));
}
-
-
-
-/*
- * Map inode to disk block and offset.
- *
- * mp -- the mount point structure for the current file system
- * tp -- the current transaction
- * ino -- the inode number of the inode to be located
- * imap -- this structure is filled in with the information necessary
- * to retrieve the given inode from disk
- * flags -- flags to pass to xfs_dilocate indicating whether or not
- * lookups in the inode btree were OK or not
- */
-int
-xfs_imap(
- xfs_mount_t *mp,
- xfs_trans_t *tp,
- xfs_ino_t ino,
- xfs_imap_t *imap,
- uint flags)
-{
- xfs_fsblock_t fsbno;
- int len;
- int off;
- int error;
-
- fsbno = imap->im_blkno ?
- XFS_DADDR_TO_FSB(mp, imap->im_blkno) : NULLFSBLOCK;
- error = xfs_dilocate(mp, tp, ino, &fsbno, &len, &off, flags);
- if (error)
- return error;
-
- imap->im_blkno = XFS_FSB_TO_DADDR(mp, fsbno);
- imap->im_len = XFS_FSB_TO_BB(mp, len);
- imap->im_agblkno = XFS_FSB_TO_AGBNO(mp, fsbno);
- imap->im_ioffset = (ushort)off;
- imap->im_boffset = (ushort)(off << mp->m_sb.sb_inodelog);
-
- /*
- * If the inode number maps to a block outside the bounds
- * of the file system then return NULL rather than calling
- * read_buf and panicing when we get an error from the
- * driver.
- */
- if ((imap->im_blkno + imap->im_len) >
- XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "(imap->im_blkno (0x%llx) + imap->im_len (0x%llx)) > "
- " XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) (0x%llx)",
- (unsigned long long) imap->im_blkno,
- (unsigned long long) imap->im_len,
- XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks));
- return EINVAL;
- }
- return 0;
-}
-
void
xfs_idestroy_fork(
xfs_inode_t *ip,
}
/*
- * This is called free all the memory associated with an inode.
- * It must free the inode itself and any buffers allocated for
- * if_extents/if_data and if_broot. It must also free the lock
- * associated with the inode.
+ * This is called to unpin an inode. The caller must have the inode locked
+ * in at least shared mode so that the buffer cannot be subsequently pinned
+ * once someone is waiting for it to be unpinned.
*/
-void
-xfs_idestroy(
- xfs_inode_t *ip)
+static void
+xfs_iunpin_nowait(
+ struct xfs_inode *ip)
{
- switch (ip->i_d.di_mode & S_IFMT) {
- case S_IFREG:
- case S_IFDIR:
- case S_IFLNK:
- xfs_idestroy_fork(ip, XFS_DATA_FORK);
- break;
- }
- if (ip->i_afp)
- xfs_idestroy_fork(ip, XFS_ATTR_FORK);
- mrfree(&ip->i_lock);
- mrfree(&ip->i_iolock);
- freesema(&ip->i_flock);
-
-#ifdef XFS_INODE_TRACE
- ktrace_free(ip->i_trace);
-#endif
-#ifdef XFS_BMAP_TRACE
- ktrace_free(ip->i_xtrace);
-#endif
-#ifdef XFS_BMBT_TRACE
- ktrace_free(ip->i_btrace);
-#endif
-#ifdef XFS_RW_TRACE
- ktrace_free(ip->i_rwtrace);
-#endif
-#ifdef XFS_ILOCK_TRACE
- ktrace_free(ip->i_lock_trace);
-#endif
-#ifdef XFS_DIR2_TRACE
- ktrace_free(ip->i_dir_trace);
-#endif
- if (ip->i_itemp) {
- /*
- * Only if we are shutting down the fs will we see an
- * inode still in the AIL. If it is there, we should remove
- * it to prevent a use-after-free from occurring.
- */
- xfs_mount_t *mp = ip->i_mount;
- xfs_log_item_t *lip = &ip->i_itemp->ili_item;
-
- ASSERT(((lip->li_flags & XFS_LI_IN_AIL) == 0) ||
- XFS_FORCED_SHUTDOWN(ip->i_mount));
- if (lip->li_flags & XFS_LI_IN_AIL) {
- spin_lock(&mp->m_ail_lock);
- if (lip->li_flags & XFS_LI_IN_AIL)
- xfs_trans_delete_ail(mp, lip);
- else
- spin_unlock(&mp->m_ail_lock);
- }
- xfs_inode_item_destroy(ip);
- }
- kmem_zone_free(xfs_inode_zone, ip);
-}
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
+ trace_xfs_inode_unpin_nowait(ip, _RET_IP_);
-/*
- * Increment the pin count of the given buffer.
- * This value is protected by ipinlock spinlock in the mount structure.
- */
-void
-xfs_ipin(
- xfs_inode_t *ip)
-{
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+ /* Give the log a push to start the unpinning I/O */
+ xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0);
- atomic_inc(&ip->i_pincount);
}
-/*
- * Decrement the pin count of the given inode, and wake up
- * anyone in xfs_iwait_unpin() if the count goes to 0. The
- * inode must have been previously pinned with a call to xfs_ipin().
- */
void
-xfs_iunpin(
- xfs_inode_t *ip)
-{
- ASSERT(atomic_read(&ip->i_pincount) > 0);
-
- if (atomic_dec_and_test(&ip->i_pincount))
- wake_up(&ip->i_ipin_wait);
-}
-
-/*
- * This is called to unpin an inode. It can be directed to wait or to return
- * immediately without waiting for the inode to be unpinned. The caller must
- * have the inode locked in at least shared mode so that the buffer cannot be
- * subsequently pinned once someone is waiting for it to be unpinned.
- */
-STATIC void
-__xfs_iunpin_wait(
- xfs_inode_t *ip,
- int wait)
-{
- xfs_inode_log_item_t *iip = ip->i_itemp;
-
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- if (atomic_read(&ip->i_pincount) == 0)
- return;
-
- /* Give the log a push to start the unpinning I/O */
- xfs_log_force(ip->i_mount, (iip && iip->ili_last_lsn) ?
- iip->ili_last_lsn : 0, XFS_LOG_FORCE);
- if (wait)
- wait_event(ip->i_ipin_wait, (atomic_read(&ip->i_pincount) == 0));
-}
-
-static inline void
xfs_iunpin_wait(
- xfs_inode_t *ip)
+ struct xfs_inode *ip)
{
- __xfs_iunpin_wait(ip, 1);
-}
-
-static inline void
-xfs_iunpin_nowait(
- xfs_inode_t *ip)
-{
- __xfs_iunpin_wait(ip, 0);
+ if (xfs_ipincount(ip)) {
+ xfs_iunpin_nowait(ip);
+ wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0));
+ }
}
-
/*
* xfs_iextents_copy()
*
for (i = 0; i < nrecs; i++) {
xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
start_block = xfs_bmbt_get_startblock(ep);
- if (ISNULLSTARTBLOCK(start_block)) {
+ if (isnullstartblock(start_block)) {
/*
* It's a delayed allocation extent, so skip it.
*/
ASSERT(ifp->if_broot_bytes <=
(XFS_IFORK_SIZE(ip, whichfork) +
XFS_BROOT_SIZE_ADJ));
- xfs_bmbt_to_bmdr(ifp->if_broot, ifp->if_broot_bytes,
+ xfs_bmbt_to_bmdr(mp, ifp->if_broot, ifp->if_broot_bytes,
(xfs_bmdr_block_t *)cp,
XFS_DFORK_SIZE(dip, mp, whichfork));
}
case XFS_DINODE_FMT_DEV:
if (iip->ili_format.ilf_fields & XFS_ILOG_DEV) {
ASSERT(whichfork == XFS_DATA_FORK);
- dip->di_u.di_dev = cpu_to_be32(ip->i_df.if_u2.if_rdev);
+ xfs_dinode_put_rdev(dip, ip->i_df.if_u2.if_rdev);
}
break;
case XFS_DINODE_FMT_UUID:
if (iip->ili_format.ilf_fields & XFS_ILOG_UUID) {
ASSERT(whichfork == XFS_DATA_FORK);
- memcpy(&dip->di_u.di_muuid, &ip->i_df.if_u2.if_uuid,
- sizeof(uuid_t));
+ memcpy(XFS_DFORK_DPTR(dip),
+ &ip->i_df.if_u2.if_uuid,
+ sizeof(uuid_t));
}
break;
xfs_buf_t *bp)
{
xfs_mount_t *mp = ip->i_mount;
- xfs_perag_t *pag = xfs_get_perag(mp, ip->i_ino);
+ struct xfs_perag *pag;
unsigned long first_index, mask;
unsigned long inodes_per_cluster;
int ilist_size;
int bufwasdelwri;
int i;
- ASSERT(pag->pagi_inodeok);
- ASSERT(pag->pag_ici_init);
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
inodes_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog;
ilist_size = inodes_per_cluster * sizeof(xfs_inode_t *);
ilist = kmem_alloc(ilist_size, KM_MAYFAIL|KM_NOFS);
if (!ilist)
- return 0;
+ goto out_put;
mask = ~(((XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog)) - 1);
first_index = XFS_INO_TO_AGINO(mp, ip->i_ino) & mask;
out_free:
read_unlock(&pag->pag_ici_lock);
kmem_free(ilist);
+out_put:
+ xfs_perag_put(pag);
return 0;
XFS_BUF_CLR_BDSTRAT_FUNC(bp);
XFS_BUF_UNDONE(bp);
XFS_BUF_STALE(bp);
- XFS_BUF_SHUT(bp);
XFS_BUF_ERROR(bp,EIO);
xfs_biodone(bp);
} else {
*/
xfs_iflush_abort(iq);
kmem_free(ilist);
+ xfs_perag_put(pag);
return XFS_ERROR(EFSCORRUPTED);
}
/*
* xfs_iflush() will write a modified inode's changes out to the
* inode's on disk home. The caller must have the inode lock held
- * in at least shared mode and the inode flush semaphore must be
- * held as well. The inode lock will still be held upon return from
+ * in at least shared mode and the inode flush completion must be
+ * active as well. The inode lock will still be held upon return from
* the call and the caller is free to unlock it.
- * The inode flush lock will be unlocked when the inode reaches the disk.
+ * The inode flush will be completed when the inode reaches the disk.
* The flags indicate how the inode's buffer should be written out.
*/
int
xfs_dinode_t *dip;
xfs_mount_t *mp;
int error;
- int noblock = (flags == XFS_IFLUSH_ASYNC_NOBLOCK);
- enum { INT_DELWRI = (1 << 0), INT_ASYNC = (1 << 1) };
XFS_STATS_INC(xs_iflush_count);
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- ASSERT(issemalocked(&(ip->i_flock)));
+ ASSERT(!completion_done(&ip->i_flush));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
mp = ip->i_mount;
/*
- * If the inode isn't dirty, then just release the inode
- * flush lock and do nothing.
- */
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- return 0;
- }
-
- /*
* We can't flush the inode until it is unpinned, so wait for it if we
* are allowed to block. We know noone new can pin it, because we are
* holding the inode lock shared and you need to hold it exclusively to
* in the same cluster are dirty, they will probably write the inode
* out for us if they occur after the log force completes.
*/
- if (noblock && xfs_ipincount(ip)) {
+ if (!(flags & SYNC_WAIT) && xfs_ipincount(ip)) {
xfs_iunpin_nowait(ip);
xfs_ifunlock(ip);
return EAGAIN;
xfs_iunpin_wait(ip);
/*
+ * For stale inodes we cannot rely on the backing buffer remaining
+ * stale in cache for the remaining life of the stale inode and so
+ * xfs_itobp() below may give us a buffer that no longer contains
+ * inodes below. We have to check this after ensuring the inode is
+ * unpinned so that it is safe to reclaim the stale inode after the
+ * flush call.
+ */
+ if (xfs_iflags_test(ip, XFS_ISTALE)) {
+ xfs_ifunlock(ip);
+ return 0;
+ }
+
+ /*
* This may have been unpinned because the filesystem is shutting
* down forcibly. If that's the case we must not write this inode
* to disk, because the log record didn't make it to disk!
}
/*
- * Decide how buffer will be flushed out. This is done before
- * the call to xfs_iflush_int because this field is zeroed by it.
- */
- if (iip != NULL && iip->ili_format.ilf_fields != 0) {
- /*
- * Flush out the inode buffer according to the directions
- * of the caller. In the cases where the caller has given
- * us a choice choose the non-delwri case. This is because
- * the inode is in the AIL and we need to get it out soon.
- */
- switch (flags) {
- case XFS_IFLUSH_SYNC:
- case XFS_IFLUSH_DELWRI_ELSE_SYNC:
- flags = 0;
- break;
- case XFS_IFLUSH_ASYNC_NOBLOCK:
- case XFS_IFLUSH_ASYNC:
- case XFS_IFLUSH_DELWRI_ELSE_ASYNC:
- flags = INT_ASYNC;
- break;
- case XFS_IFLUSH_DELWRI:
- flags = INT_DELWRI;
- break;
- default:
- ASSERT(0);
- flags = 0;
- break;
- }
- } else {
- switch (flags) {
- case XFS_IFLUSH_DELWRI_ELSE_SYNC:
- case XFS_IFLUSH_DELWRI_ELSE_ASYNC:
- case XFS_IFLUSH_DELWRI:
- flags = INT_DELWRI;
- break;
- case XFS_IFLUSH_ASYNC_NOBLOCK:
- case XFS_IFLUSH_ASYNC:
- flags = INT_ASYNC;
- break;
- case XFS_IFLUSH_SYNC:
- flags = 0;
- break;
- default:
- ASSERT(0);
- flags = 0;
- break;
- }
- }
-
- /*
* Get the buffer containing the on-disk inode.
*/
- error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0, 0,
- noblock ? XFS_BUF_TRYLOCK : XFS_BUF_LOCK);
+ error = xfs_itobp(mp, NULL, ip, &dip, &bp,
+ (flags & SYNC_WAIT) ? XBF_LOCK : XBF_TRYLOCK);
if (error || !bp) {
xfs_ifunlock(ip);
return error;
* get stuck waiting in the write for too long.
*/
if (XFS_BUF_ISPINNED(bp))
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_log_force(mp, 0);
/*
* inode clustering:
if (error)
goto cluster_corrupt_out;
- if (flags & INT_DELWRI) {
- xfs_bdwrite(mp, bp);
- } else if (flags & INT_ASYNC) {
- error = xfs_bawrite(mp, bp);
- } else {
+ if (flags & SYNC_WAIT)
error = xfs_bwrite(mp, bp);
- }
+ else
+ xfs_bdwrite(mp, bp);
return error;
corrupt_out:
#endif
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- ASSERT(issemalocked(&(ip->i_flock)));
+ ASSERT(!completion_done(&ip->i_flush));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
iip = ip->i_itemp;
mp = ip->i_mount;
-
- /*
- * If the inode isn't dirty, then just release the inode
- * flush lock and do nothing.
- */
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- return 0;
- }
-
/* set *dip = inode's place in the buffer */
- dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_boffset);
+ dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
/*
* Clear i_update_core before copying out the data.
SYNCHRONIZE();
/*
- * Make sure to get the latest atime from the Linux inode.
+ * Make sure to get the latest timestamps from the Linux inode.
*/
- xfs_synchronize_atime(ip);
+ xfs_synchronize_times(ip);
- if (XFS_TEST_ERROR(be16_to_cpu(dip->di_core.di_magic) != XFS_DINODE_MAGIC,
+ if (XFS_TEST_ERROR(be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC,
mp, XFS_ERRTAG_IFLUSH_1, XFS_RANDOM_IFLUSH_1)) {
xfs_cmn_err(XFS_PTAG_IFLUSH, CE_ALERT, mp,
"xfs_iflush: Bad inode %Lu magic number 0x%x, ptr 0x%p",
- ip->i_ino, be16_to_cpu(dip->di_core.di_magic), dip);
+ ip->i_ino, be16_to_cpu(dip->di_magic), dip);
goto corrupt_out;
}
if (XFS_TEST_ERROR(ip->i_d.di_magic != XFS_DINODE_MAGIC,
* because if the inode is dirty at all the core must
* be.
*/
- xfs_dinode_to_disk(&dip->di_core, &ip->i_d);
+ xfs_dinode_to_disk(dip, &ip->i_d);
/* Wrap, we never let the log put out DI_MAX_FLUSH */
if (ip->i_d.di_flushiter == DI_MAX_FLUSH)
* convert back to the old inode format. If the superblock version
* has been updated, then make the conversion permanent.
*/
- ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 ||
- xfs_sb_version_hasnlink(&mp->m_sb));
- if (ip->i_d.di_version == XFS_DINODE_VERSION_1) {
+ ASSERT(ip->i_d.di_version == 1 || xfs_sb_version_hasnlink(&mp->m_sb));
+ if (ip->i_d.di_version == 1) {
if (!xfs_sb_version_hasnlink(&mp->m_sb)) {
/*
* Convert it back.
*/
ASSERT(ip->i_d.di_nlink <= XFS_MAXLINK_1);
- dip->di_core.di_onlink = cpu_to_be16(ip->i_d.di_nlink);
+ dip->di_onlink = cpu_to_be16(ip->i_d.di_nlink);
} else {
/*
* The superblock version has already been bumped,
* so just make the conversion to the new inode
* format permanent.
*/
- ip->i_d.di_version = XFS_DINODE_VERSION_2;
- dip->di_core.di_version = XFS_DINODE_VERSION_2;
+ ip->i_d.di_version = 2;
+ dip->di_version = 2;
ip->i_d.di_onlink = 0;
- dip->di_core.di_onlink = 0;
+ dip->di_onlink = 0;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
- memset(&(dip->di_core.di_pad[0]), 0,
- sizeof(dip->di_core.di_pad));
+ memset(&(dip->di_pad[0]), 0,
+ sizeof(dip->di_pad));
ASSERT(ip->i_d.di_projid == 0);
}
}
iip->ili_format.ilf_fields = 0;
iip->ili_logged = 1;
- ASSERT(sizeof(xfs_lsn_t) == 8); /* don't lock if it shrinks */
- spin_lock(&mp->m_ail_lock);
- iip->ili_flush_lsn = iip->ili_item.li_lsn;
- spin_unlock(&mp->m_ail_lock);
+ xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
+ &iip->ili_item.li_lsn);
/*
* Attach the function xfs_iflush_done to the inode's
return XFS_ERROR(EFSCORRUPTED);
}
-
-/*
- * Flush all inactive inodes in mp.
- */
-void
-xfs_iflush_all(
- xfs_mount_t *mp)
-{
- xfs_inode_t *ip;
-
- again:
- XFS_MOUNT_ILOCK(mp);
- ip = mp->m_inodes;
- if (ip == NULL)
- goto out;
-
- do {
- /* Make sure we skip markers inserted by sync */
- if (ip->i_mount == NULL) {
- ip = ip->i_mnext;
- continue;
- }
-
- if (!VFS_I(ip)) {
- XFS_MOUNT_IUNLOCK(mp);
- xfs_finish_reclaim(ip, 0, XFS_IFLUSH_ASYNC);
- goto again;
- }
-
- ASSERT(vn_count(VFS_I(ip)) == 0);
-
- ip = ip->i_mnext;
- } while (ip != mp->m_inodes);
- out:
- XFS_MOUNT_IUNLOCK(mp);
-}
-
-#ifdef XFS_ILOCK_TRACE
-ktrace_t *xfs_ilock_trace_buf;
-
-void
-xfs_ilock_trace(xfs_inode_t *ip, int lock, unsigned int lockflags, inst_t *ra)
-{
- ktrace_enter(ip->i_lock_trace,
- (void *)ip,
- (void *)(unsigned long)lock, /* 1 = LOCK, 3=UNLOCK, etc */
- (void *)(unsigned long)lockflags, /* XFS_ILOCK_EXCL etc */
- (void *)ra, /* caller of ilock */
- (void *)(unsigned long)current_cpu(),
- (void *)(unsigned long)current_pid(),
- NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL);
-}
-#endif
-
/*
* Return a pointer to the extent record at file index idx.
*/
*/
void
xfs_iext_insert(
- xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* starting index of new items */
xfs_extnum_t count, /* number of inserted items */
- xfs_bmbt_irec_t *new) /* items to insert */
+ xfs_bmbt_irec_t *new, /* items to insert */
+ int state) /* type of extent conversion */
{
+ xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df;
xfs_extnum_t i; /* extent record index */
+ trace_xfs_iext_insert(ip, idx, new, state, _RET_IP_);
+
ASSERT(ifp->if_flags & XFS_IFEXTENTS);
xfs_iext_add(ifp, idx, count);
for (i = idx; i < idx + count; i++, new++)
*/
void
xfs_iext_remove(
- xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* index to begin removing exts */
- int ext_diff) /* number of extents to remove */
+ int ext_diff, /* number of extents to remove */
+ int state) /* type of extent conversion */
{
+ xfs_ifork_t *ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df;
xfs_extnum_t nextents; /* number of extents in file */
int new_size; /* size of extents after removal */
+ trace_xfs_iext_remove(ip, idx, state, _RET_IP_);
+
ASSERT(ext_diff > 0);
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
new_size = (nextents - ext_diff) * sizeof(xfs_bmbt_rec_t);
/*
* Resize an extent indirection array to new_size bytes.
*/
-void
+STATIC void
xfs_iext_realloc_indirect(
xfs_ifork_t *ifp, /* inode fork pointer */
int new_size) /* new indirection array size */
/*
* Switch from indirection array to linear (direct) extent allocations.
*/
-void
+STATIC void
xfs_iext_indirect_to_direct(
xfs_ifork_t *ifp) /* inode fork pointer */
{
ASSERT(nextents <= XFS_LINEAR_EXTS);
size = nextents * sizeof(xfs_bmbt_rec_t);
- xfs_iext_irec_compact_full(ifp);
+ xfs_iext_irec_compact_pages(ifp);
ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ);
ep = ifp->if_u1.if_ext_irec->er_extbuf;
* compaction policy is as follows:
*
* Full Compaction: Extents fit into a single page (or inline buffer)
- * Full Compaction: Extents occupy less than 10% of allocated space
- * Partial Compaction: Extents occupy > 10% and < 50% of allocated space
+ * Partial Compaction: Extents occupy less than 50% of allocated space
* No Compaction: Extents occupy at least 50% of allocated space
*/
void
xfs_iext_direct_to_inline(ifp, nextents);
} else if (nextents <= XFS_LINEAR_EXTS) {
xfs_iext_indirect_to_direct(ifp);
- } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 3) {
- xfs_iext_irec_compact_full(ifp);
} else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) {
xfs_iext_irec_compact_pages(ifp);
}
erp_next = erp + 1;
if (erp_next->er_extcount <=
(XFS_LINEAR_EXTS - erp->er_extcount)) {
- memmove(&erp->er_extbuf[erp->er_extcount],
+ memcpy(&erp->er_extbuf[erp->er_extcount],
erp_next->er_extbuf, erp_next->er_extcount *
sizeof(xfs_bmbt_rec_t));
erp->er_extcount += erp_next->er_extcount;
}
/*
- * Fully compact the extent records managed by the indirection array.
- */
-void
-xfs_iext_irec_compact_full(
- xfs_ifork_t *ifp) /* inode fork pointer */
-{
- xfs_bmbt_rec_host_t *ep, *ep_next; /* extent record pointers */
- xfs_ext_irec_t *erp, *erp_next; /* extent irec pointers */
- int erp_idx = 0; /* extent irec index */
- int ext_avail; /* empty entries in ex list */
- int ext_diff; /* number of exts to add */
- int nlists; /* number of irec's (ex lists) */
-
- ASSERT(ifp->if_flags & XFS_IFEXTIREC);
-
- nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
- erp = ifp->if_u1.if_ext_irec;
- ep = &erp->er_extbuf[erp->er_extcount];
- erp_next = erp + 1;
- ep_next = erp_next->er_extbuf;
-
- while (erp_idx < nlists - 1) {
- /*
- * Check how many extent records are available in this irec.
- * If there is none skip the whole exercise.
- */
- ext_avail = XFS_LINEAR_EXTS - erp->er_extcount;
- if (ext_avail) {
-
- /*
- * Copy over as many as possible extent records into
- * the previous page.
- */
- ext_diff = MIN(ext_avail, erp_next->er_extcount);
- memcpy(ep, ep_next, ext_diff * sizeof(xfs_bmbt_rec_t));
- erp->er_extcount += ext_diff;
- erp_next->er_extcount -= ext_diff;
-
- /*
- * If the next irec is empty now we can simply
- * remove it.
- */
- if (erp_next->er_extcount == 0) {
- /*
- * Free page before removing extent record
- * so er_extoffs don't get modified in
- * xfs_iext_irec_remove.
- */
- kmem_free(erp_next->er_extbuf);
- erp_next->er_extbuf = NULL;
- xfs_iext_irec_remove(ifp, erp_idx + 1);
- erp = &ifp->if_u1.if_ext_irec[erp_idx];
- nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
-
- /*
- * If the next irec is not empty move up the content
- * that has not been copied to the previous page to
- * the beggining of this one.
- */
- } else {
- memmove(erp_next->er_extbuf, &ep_next[ext_diff],
- erp_next->er_extcount *
- sizeof(xfs_bmbt_rec_t));
- ep_next = erp_next->er_extbuf;
- memset(&ep_next[erp_next->er_extcount], 0,
- (XFS_LINEAR_EXTS -
- erp_next->er_extcount) *
- sizeof(xfs_bmbt_rec_t));
- }
- }
-
- if (erp->er_extcount == XFS_LINEAR_EXTS) {
- erp_idx++;
- if (erp_idx < nlists)
- erp = &ifp->if_u1.if_ext_irec[erp_idx];
- else
- break;
- }
- ep = &erp->er_extbuf[erp->er_extcount];
- erp_next = erp + 1;
- ep_next = erp_next->er_extbuf;
- }
-}
-
-/*
* This is called to update the er_extoff field in the indirection
* array when extents have been added or removed from one of the
* extent lists. erp_idx contains the irec index to begin updating