static inline bool
xfs_buf_item_straddle(
struct xfs_buf *bp,
uint offset,
int next_bit,
int last_bit)
{
return xfs_buf_offset(bp, offset + (next_bit << XFS_BLF_SHIFT)) !=
(xfs_buf_offset(bp, offset + (last_bit << XFS_BLF_SHIFT)) +
XFS_BLF_CHUNK);
}
/*
* This returns the number of log iovecs needed to log the
* given buf log item.
*
* It calculates this as 1 iovec for the buf log format structure
* and 1 for each stretch of non-contiguous chunks to be logged.
* Contiguous chunks are logged in a single iovec.
*
* If the XFS_BLI_STALE flag has been set, then log nothing.
*/
STATIC void
xfs_buf_item_size_segment(
struct xfs_buf_log_item *bip,
struct xfs_buf_log_format *blfp,
int *nvecs,
int *nbytes)
{
struct xfs_buf *bp = bip->bli_buf;
int next_bit;
int last_bit;
last_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0);
if (last_bit == -1)
return;
/*
* initial count for a dirty buffer is 2 vectors - the format structure
* and the first dirty region.
*/
*nvecs += 2;
*nbytes += xfs_buf_log_format_size(blfp) + XFS_BLF_CHUNK;
while (last_bit != -1) {
/*
* This takes the bit number to start looking from and
* returns the next set bit from there. It returns -1
* if there are no more bits set or the start bit is
* beyond the end of the bitmap.
*/
next_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size,
last_bit + 1);
/*
* If we run out of bits, leave the loop,
* else if we find a new set of bits bump the number of vecs,
* else keep scanning the current set of bits.
*/
if (next_bit == -1) {
break;
} else if (next_bit != last_bit + 1) {
last_bit = next_bit;
(*nvecs)++;
} else if (xfs_buf_offset(bp, next_bit * XFS_BLF_CHUNK) !=
(xfs_buf_offset(bp, last_bit * XFS_BLF_CHUNK) +
XFS_BLF_CHUNK)) {
last_bit = next_bit;
(*nvecs)++;
} else {
last_bit++;
}
*nbytes += XFS_BLF_CHUNK;
}
}
/*
* This routine is called to map an inode to the buffer containing the on-disk
* version of the inode. It returns a pointer to the buffer containing the
* on-disk inode in the bpp parameter, and in the dipp parameter it returns a
* pointer to the on-disk inode within that buffer.
*
* If a non-zero error is returned, then the contents of bpp and dipp are
* undefined.
*/
int
xfs_imap_to_bp(
struct xfs_mount *mp,
struct xfs_trans *tp,
struct xfs_imap *imap,
struct xfs_dinode **dipp,
struct xfs_buf **bpp,
uint buf_flags,
uint iget_flags)
{
struct xfs_buf *bp;
int error;
buf_flags |= XBF_UNMAPPED;
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno,
(int)imap->im_len, buf_flags, &bp,
&xfs_inode_buf_ops);
if (error) {
if (error == -EAGAIN) {
ASSERT(buf_flags & XBF_TRYLOCK);
return error;
}
xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.",
__func__, error);
return error;
}
*bpp = bp;
*dipp = xfs_buf_offset(bp, imap->im_boffset);
return 0;
}
/*
* Writes a modified inode's changes out to the inode's on disk home.
* Originally based on xfs_iflush_int() from xfs_inode.c in the kernel.
*/
int
libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
{
xfs_inode_log_item_t *iip;
xfs_dinode_t *dip;
xfs_mount_t *mp;
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
ASSERT(ip->i_d.di_version > 1);
iip = ip->i_itemp;
mp = ip->i_mount;
/* set *dip = inode's place in the buffer */
dip = xfs_buf_offset(bp, ip->i_imap.im_boffset);
ASSERT(ip->i_d.di_magic == XFS_DINODE_MAGIC);
if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) {
ASSERT( (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS) ||
(ip->i_d.di_format == XFS_DINODE_FMT_BTREE) );
}
else if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
ASSERT( (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS) ||
(ip->i_d.di_format == XFS_DINODE_FMT_BTREE) ||
(ip->i_d.di_format == XFS_DINODE_FMT_LOCAL) );
}
ASSERT(ip->i_d.di_nextents+ip->i_d.di_anextents <= ip->i_d.di_nblocks);
ASSERT(ip->i_d.di_forkoff <= mp->m_sb.sb_inodesize);
/* bump the change count on v3 inodes */
if (ip->i_d.di_version == 3)
ip->i_d.di_changecount++;
/*
* Copy the dirty parts of the inode into the on-disk
* inode. We always copy out the core of the inode,
* because if the inode is dirty at all the core must
* be.
*/
xfs_dinode_to_disk(dip, &ip->i_d);
xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK);
if (XFS_IFORK_Q(ip))
xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK);
/* update the lsn in the on disk inode if required */
if (ip->i_d.di_version == 3)
dip->di_lsn = cpu_to_be64(iip->ili_item.li_lsn);
/* generate the checksum. */
xfs_dinode_calc_crc(mp, dip);
return 0;
}
/* Check a particular inode with ir_free. */
STATIC int
xfs_scrub_iallocbt_check_cluster_freemask(
struct xfs_scrub_btree *bs,
xfs_ino_t fsino,
xfs_agino_t chunkino,
xfs_agino_t clusterino,
struct xfs_inobt_rec_incore *irec,
struct xfs_buf *bp)
{
struct xfs_dinode *dip;
struct xfs_mount *mp = bs->cur->bc_mp;
bool inode_is_free = false;
bool freemask_ok;
bool inuse;
int error = 0;
if (xfs_scrub_should_terminate(bs->sc, &error))
return error;
dip = xfs_buf_offset(bp, clusterino * mp->m_sb.sb_inodesize);
if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC ||
(dip->di_version >= 3 &&
be64_to_cpu(dip->di_ino) != fsino + clusterino)) {
xfs_scrub_btree_set_corrupt(bs->sc, bs->cur, 0);
goto out;
}
if (irec->ir_free & XFS_INOBT_MASK(chunkino + clusterino))
inode_is_free = true;
error = xfs_icache_inode_is_allocated(mp, bs->cur->bc_tp,
fsino + clusterino, &inuse);
if (error == -ENODATA) {
/* Not cached, just read the disk buffer */
freemask_ok = inode_is_free ^ !!(dip->di_mode);
if (!bs->sc->try_harder && !freemask_ok)
return -EDEADLOCK;
} else if (error < 0) {
/*
* Inode is only half assembled, or there was an IO error,
* or the verifier failed, so don't bother trying to check.
* The inode scrubber can deal with this.
*/
goto out;
} else {
/* Inode is all there. */
freemask_ok = inode_is_free ^ inuse;
}
if (!freemask_ok)
xfs_scrub_btree_set_corrupt(bs->sc, bs->cur, 0);
out:
return 0;
}
static inline void
xfs_buf_item_copy_iovec(
struct xfs_log_vec *lv,
struct xfs_log_iovec **vecp,
struct xfs_buf *bp,
uint offset,
int first_bit,
uint nbits)
{
offset += first_bit * XFS_BLF_CHUNK;
xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_BCHUNK,
xfs_buf_offset(bp, offset),
nbits * XFS_BLF_CHUNK);
}
/*
* If we are doing readahead on an inode buffer, we might be in log recovery
* reading an inode allocation buffer that hasn't yet been replayed, and hence
* has not had the inode cores stamped into it. Hence for readahead, the buffer
* may be potentially invalid.
*
* If the readahead buffer is invalid, we need to mark it with an error and
* clear the DONE status of the buffer so that a followup read will re-read it
* from disk. We don't report the error otherwise to avoid warnings during log
* recovery and we don't get unnecssary panics on debug kernels. We use EIO here
* because all we want to do is say readahead failed; there is no-one to report
* the error to, so this will distinguish it from a non-ra verifier failure.
* Changes to this readahead error behavour also need to be reflected in
* xfs_dquot_buf_readahead_verify().
*/
static void
xfs_inode_buf_verify(
struct xfs_buf *bp,
bool readahead)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
xfs_agnumber_t agno;
int i;
int ni;
/*
* Validate the magic number and version of every inode in the buffer
*/
agno = xfs_daddr_to_agno(mp, XFS_BUF_ADDR(bp));
ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
xfs_agino_t unlinked_ino;
dip = xfs_buf_offset(bp, (i << mp->m_sb.sb_inodelog));
unlinked_ino = be32_to_cpu(dip->di_next_unlinked);
di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) &&
xfs_dinode_good_version(mp, dip->di_version) &&
(unlinked_ino == NULLAGINO ||
xfs_verify_agino(mp, agno, unlinked_ino));
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP))) {
if (readahead) {
bp->b_flags &= ~XBF_DONE;
xfs_buf_ioerror(bp, -EIO);
return;
}
#ifdef DEBUG
xfs_alert(mp,
"bad inode magic/vsn daddr %lld #%d (magic=%x)",
(unsigned long long)bp->b_bn, i,
be16_to_cpu(dip->di_magic));
#endif
xfs_buf_verifier_error(bp, -EFSCORRUPTED,
__func__, dip, sizeof(*dip),
NULL);
return;
}
}
}
void
xfs_inobp_check(
xfs_mount_t *mp,
xfs_buf_t *bp)
{
int i;
int j;
xfs_dinode_t *dip;
j = mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog;
for (i = 0; i < j; i++) {
dip = xfs_buf_offset(bp, i * mp->m_sb.sb_inodesize);
if (!dip->di_next_unlinked) {
xfs_alert(mp,
"Detected bogus zero next_unlinked field in inode %d buffer 0x%llx.",
i, (long long)bp->b_bn);
}
}
}
/*
* If we are doing readahead on an inode buffer, we might be in log recovery
* reading an inode allocation buffer that hasn't yet been replayed, and hence
* has not had the inode cores stamped into it. Hence for readahead, the buffer
* may be potentially invalid.
*
* If the readahead buffer is invalid, we don't want to mark it with an error,
* but we do want to clear the DONE status of the buffer so that a followup read
* will re-read it from disk. This will ensure that we don't get an unnecessary
* warnings during log recovery and we don't get unnecssary panics on debug
* kernels.
*/
static void
xfs_inode_buf_verify(
struct xfs_buf *bp,
bool readahead)
{
struct xfs_mount *mp = bp->b_target->bt_mount;
int i;
int ni;
/*
* Validate the magic number and version of every inode in the buffer
*/
ni = XFS_BB_TO_FSB(mp, bp->b_length) * mp->m_sb.sb_inopblock;
for (i = 0; i < ni; i++) {
int di_ok;
xfs_dinode_t *dip;
dip = (struct xfs_dinode *)xfs_buf_offset(bp,
(i << mp->m_sb.sb_inodelog));
di_ok = dip->di_magic == cpu_to_be16(XFS_DINODE_MAGIC) &&
XFS_DINODE_GOOD_VERSION(dip->di_version);
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
if (readahead) {
bp->b_flags &= ~XBF_DONE;
return;
}
xfs_buf_ioerror(bp, -EFSCORRUPTED);
xfs_verifier_error(bp);
#ifdef DEBUG
xfs_alert(mp,
"bad inode magic/vsn daddr %lld #%d (magic=%x)",
(unsigned long long)bp->b_bn, i,
be16_to_cpu(dip->di_magic));
#endif
}
}
xfs_inobp_check(mp, bp);
}
/*
* Check that the holemask and freemask of a hypothetical inode cluster match
* what's actually on disk. If sparse inodes are enabled, the cluster does
* not actually have to map to inodes if the corresponding holemask bit is set.
*
* @cluster_base is the first inode in the cluster within the @irec.
*/
STATIC int
xchk_iallocbt_check_cluster(
struct xchk_btree *bs,
struct xfs_inobt_rec_incore *irec,
unsigned int cluster_base)
{
struct xfs_imap imap;
struct xfs_mount *mp = bs->cur->bc_mp;
struct xfs_dinode *dip;
struct xfs_buf *cluster_bp;
unsigned int nr_inodes;
xfs_agnumber_t agno = bs->cur->bc_private.a.agno;
xfs_agblock_t agbno;
unsigned int cluster_index;
uint16_t cluster_mask = 0;
uint16_t ir_holemask;
int error = 0;
nr_inodes = min_t(unsigned int, XFS_INODES_PER_CHUNK,
mp->m_inodes_per_cluster);
/* Map this inode cluster */
agbno = XFS_AGINO_TO_AGBNO(mp, irec->ir_startino + cluster_base);
/* Compute a bitmask for this cluster that can be used for holemask. */
for (cluster_index = 0;
cluster_index < nr_inodes;
cluster_index += XFS_INODES_PER_HOLEMASK_BIT)
cluster_mask |= XFS_INOBT_MASK((cluster_base + cluster_index) /
XFS_INODES_PER_HOLEMASK_BIT);
/*
* Map the first inode of this cluster to a buffer and offset.
* Be careful about inobt records that don't align with the start of
* the inode buffer when block sizes are large enough to hold multiple
* inode chunks. When this happens, cluster_base will be zero but
* ir_startino can be large enough to make im_boffset nonzero.
*/
ir_holemask = (irec->ir_holemask & cluster_mask);
imap.im_blkno = XFS_AGB_TO_DADDR(mp, agno, agbno);
imap.im_len = XFS_FSB_TO_BB(mp, mp->m_blocks_per_cluster);
imap.im_boffset = XFS_INO_TO_OFFSET(mp, irec->ir_startino);
if (imap.im_boffset != 0 && cluster_base != 0) {
ASSERT(imap.im_boffset == 0 || cluster_base == 0);
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
return 0;
}
trace_xchk_iallocbt_check_cluster(mp, agno, irec->ir_startino,
imap.im_blkno, imap.im_len, cluster_base, nr_inodes,
cluster_mask, ir_holemask,
XFS_INO_TO_OFFSET(mp, irec->ir_startino +
cluster_base));
/* The whole cluster must be a hole or not a hole. */
if (ir_holemask != cluster_mask && ir_holemask != 0) {
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
return 0;
}
/* If any part of this is a hole, skip it. */
if (ir_holemask) {
xchk_xref_is_not_owned_by(bs->sc, agbno,
mp->m_blocks_per_cluster,
&XFS_RMAP_OINFO_INODES);
return 0;
}
xchk_xref_is_owned_by(bs->sc, agbno, mp->m_blocks_per_cluster,
&XFS_RMAP_OINFO_INODES);
/* Grab the inode cluster buffer. */
error = xfs_imap_to_bp(mp, bs->cur->bc_tp, &imap, &dip, &cluster_bp,
0, 0);
if (!xchk_btree_xref_process_error(bs->sc, bs->cur, 0, &error))
return error;
/* Check free status of each inode within this cluster. */
for (cluster_index = 0; cluster_index < nr_inodes; cluster_index++) {
struct xfs_dinode *dip;
if (imap.im_boffset >= BBTOB(cluster_bp->b_length)) {
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
break;
}
dip = xfs_buf_offset(cluster_bp, imap.im_boffset);
error = xchk_iallocbt_check_cluster_ifree(bs, irec,
cluster_base + cluster_index, dip);
if (error)
break;
imap.im_boffset += mp->m_sb.sb_inodesize;
}
xfs_trans_brelse(bs->cur->bc_tp, cluster_bp);
return error;
}
static struct xfs_log_iovec *
xfs_buf_item_format_segment(
struct xfs_buf_log_item *bip,
struct xfs_log_iovec *vecp,
uint offset,
struct xfs_buf_log_format *blfp)
{
struct xfs_buf *bp = bip->bli_buf;
uint base_size;
uint nvecs;
int first_bit;
int last_bit;
int next_bit;
uint nbits;
uint buffer_offset;
/* copy the flags across from the base format item */
blfp->blf_flags = bip->__bli_format.blf_flags;
/*
* Base size is the actual size of the ondisk structure - it reflects
* the actual size of the dirty bitmap rather than the size of the in
* memory structure.
*/
base_size = offsetof(struct xfs_buf_log_format, blf_data_map) +
(blfp->blf_map_size * sizeof(blfp->blf_data_map[0]));
nvecs = 0;
first_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0);
if (!(bip->bli_flags & XFS_BLI_STALE) && first_bit == -1) {
/*
* If the map is not be dirty in the transaction, mark
* the size as zero and do not advance the vector pointer.
*/
goto out;
}
vecp->i_addr = blfp;
vecp->i_len = base_size;
vecp->i_type = XLOG_REG_TYPE_BFORMAT;
vecp++;
nvecs = 1;
if (bip->bli_flags & XFS_BLI_STALE) {
/*
* The buffer is stale, so all we need to log
* is the buf log format structure with the
* cancel flag in it.
*/
trace_xfs_buf_item_format_stale(bip);
ASSERT(blfp->blf_flags & XFS_BLF_CANCEL);
goto out;
}
/*
* Fill in an iovec for each set of contiguous chunks.
*/
last_bit = first_bit;
nbits = 1;
for (;;) {
/*
* This takes the bit number to start looking from and
* returns the next set bit from there. It returns -1
* if there are no more bits set or the start bit is
* beyond the end of the bitmap.
*/
next_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size,
(uint)last_bit + 1);
/*
* If we run out of bits fill in the last iovec and get
* out of the loop.
* Else if we start a new set of bits then fill in the
* iovec for the series we were looking at and start
* counting the bits in the new one.
* Else we're still in the same set of bits so just
* keep counting and scanning.
*/
if (next_bit == -1) {
buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
break;
} else if (next_bit != last_bit + 1) {
buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
} else if (xfs_buf_offset(bp, offset +
(next_bit << XFS_BLF_SHIFT)) !=
(xfs_buf_offset(bp, offset +
(last_bit << XFS_BLF_SHIFT)) +
XFS_BLF_CHUNK)) {
buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
/*
* You would think we need to bump the nvecs here too, but we do not
* this number is used by recovery, and it gets confused by the boundary
* split here
* nvecs++;
*/
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
} else {
last_bit++;
nbits++;
}
}
out:
blfp->blf_size = nvecs;
return vecp;
}
/*
* This is called to fill in the vector of log iovecs for the
* given log buf item. It fills the first entry with a buf log
* format structure, and the rest point to contiguous chunks
* within the buffer.
*/
void
xfs_buf_item_format(
xfs_buf_log_item_t *bip,
xfs_log_iovec_t *log_vector)
{
uint base_size;
uint nvecs;
xfs_log_iovec_t *vecp;
xfs_buf_t *bp;
int first_bit;
int last_bit;
int next_bit;
uint nbits;
uint buffer_offset;
ASSERT(atomic_read(&bip->bli_refcount) > 0);
ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
(bip->bli_flags & XFS_BLI_STALE));
bp = bip->bli_buf;
ASSERT(XFS_BUF_BP_ISMAPPED(bp));
vecp = log_vector;
/*
* The size of the base structure is the size of the
* declared structure plus the space for the extra words
* of the bitmap. We subtract one from the map size, because
* the first element of the bitmap is accounted for in the
* size of the base structure.
*/
base_size =
(uint)(sizeof(xfs_buf_log_format_t) +
((bip->bli_format.blf_map_size - 1) * sizeof(uint)));
vecp->i_addr = (xfs_caddr_t)&bip->bli_format;
vecp->i_len = base_size;
vecp++;
nvecs = 1;
if (bip->bli_flags & XFS_BLI_STALE) {
/*
* The buffer is stale, so all we need to log
* is the buf log format structure with the
* cancel flag in it.
*/
xfs_buf_item_trace("FORMAT STALE", bip);
ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
bip->bli_format.blf_size = nvecs;
return;
}
/*
* Fill in an iovec for each set of contiguous chunks.
*/
first_bit = xfs_buf_item_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size, 0);
ASSERT(first_bit != -1);
last_bit = first_bit;
nbits = 1;
for (;;) {
/*
* This takes the bit number to start looking from and
* returns the next set bit from there. It returns -1
* if there are no more bits set or the start bit is
* beyond the end of the bitmap.
*/
next_bit = xfs_buf_item_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size,
(uint)last_bit + 1);
/*
* If we run out of bits fill in the last iovec and get
* out of the loop.
* Else if we start a new set of bits then fill in the
* iovec for the series we were looking at and start
* counting the bits in the new one.
* Else we're still in the same set of bits so just
* keep counting and scanning.
*/
if (next_bit == -1) {
buffer_offset = first_bit * XFS_BLI_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLI_CHUNK;
nvecs++;
break;
} else if (next_bit != last_bit + 1) {
buffer_offset = first_bit * XFS_BLI_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLI_CHUNK;
nvecs++;
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
} else if (xfs_buf_offset(bp, next_bit * XFS_BLI_CHUNK) !=
(xfs_buf_offset(bp, last_bit * XFS_BLI_CHUNK) +
XFS_BLI_CHUNK)) {
buffer_offset = first_bit * XFS_BLI_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLI_CHUNK;
nvecs++;
vecp++;
first_bit = next_bit;
last_bit = next_bit;
} else {
last_bit++;
nbits++;
}
}
bip->bli_format.blf_size = nvecs;
/*
* Check to make sure everything is consistent.
*/
xfs_buf_item_trace("FORMAT NORM", bip);
xfs_buf_item_log_check(bip);
}
/*
* This returns the number of log iovecs needed to log the
* given buf log item.
*
* It calculates this as 1 iovec for the buf log format structure
* and 1 for each stretch of non-contiguous chunks to be logged.
* Contiguous chunks are logged in a single iovec.
*
* If the XFS_BLI_STALE flag has been set, then log nothing.
*/
uint
xfs_buf_item_size(
xfs_buf_log_item_t *bip)
{
uint nvecs;
int next_bit;
int last_bit;
xfs_buf_t *bp;
ASSERT(atomic_read(&bip->bli_refcount) > 0);
if (bip->bli_flags & XFS_BLI_STALE) {
/*
* The buffer is stale, so all we need to log
* is the buf log format structure with the
* cancel flag in it.
*/
xfs_buf_item_trace("SIZE STALE", bip);
ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
return 1;
}
bp = bip->bli_buf;
ASSERT(bip->bli_flags & XFS_BLI_LOGGED);
nvecs = 1;
last_bit = xfs_buf_item_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size, 0);
ASSERT(last_bit != -1);
nvecs++;
while (last_bit != -1) {
/*
* This takes the bit number to start looking from and
* returns the next set bit from there. It returns -1
* if there are no more bits set or the start bit is
* beyond the end of the bitmap.
*/
next_bit = xfs_buf_item_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size,
last_bit + 1);
/*
* If we run out of bits, leave the loop,
* else if we find a new set of bits bump the number of vecs,
* else keep scanning the current set of bits.
*/
if (next_bit == -1) {
last_bit = -1;
} else if (next_bit != last_bit + 1) {
last_bit = next_bit;
nvecs++;
} else if (xfs_buf_offset(bp, next_bit * XFS_BLI_CHUNK) !=
(xfs_buf_offset(bp, last_bit * XFS_BLI_CHUNK) +
XFS_BLI_CHUNK)) {
last_bit = next_bit;
nvecs++;
} else {
last_bit++;
}
}
xfs_buf_item_trace("SIZE NORM", bip);
return nvecs;
}
/*
* This is called to fill in the vector of log iovecs for the
* given log buf item. It fills the first entry with a buf log
* format structure, and the rest point to contiguous chunks
* within the buffer.
*/
STATIC void
xfs_buf_item_format(
struct xfs_log_item *lip,
struct xfs_log_iovec *vecp)
{
struct xfs_buf_log_item *bip = BUF_ITEM(lip);
struct xfs_buf *bp = bip->bli_buf;
uint base_size;
uint nvecs;
int first_bit;
int last_bit;
int next_bit;
uint nbits;
uint buffer_offset;
ASSERT(atomic_read(&bip->bli_refcount) > 0);
ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
(bip->bli_flags & XFS_BLI_STALE));
/*
* The size of the base structure is the size of the
* declared structure plus the space for the extra words
* of the bitmap. We subtract one from the map size, because
* the first element of the bitmap is accounted for in the
* size of the base structure.
*/
base_size =
(uint)(sizeof(xfs_buf_log_format_t) +
((bip->bli_format.blf_map_size - 1) * sizeof(uint)));
vecp->i_addr = &bip->bli_format;
vecp->i_len = base_size;
vecp->i_type = XLOG_REG_TYPE_BFORMAT;
vecp++;
nvecs = 1;
/*
* If it is an inode buffer, transfer the in-memory state to the
* format flags and clear the in-memory state. We do not transfer
* this state if the inode buffer allocation has not yet been committed
* to the log as setting the XFS_BLI_INODE_BUF flag will prevent
* correct replay of the inode allocation.
*/
if (bip->bli_flags & XFS_BLI_INODE_BUF) {
if (!((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
xfs_log_item_in_current_chkpt(lip)))
bip->bli_format.blf_flags |= XFS_BLF_INODE_BUF;
bip->bli_flags &= ~XFS_BLI_INODE_BUF;
}
if (bip->bli_flags & XFS_BLI_STALE) {
/*
* The buffer is stale, so all we need to log
* is the buf log format structure with the
* cancel flag in it.
*/
trace_xfs_buf_item_format_stale(bip);
ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
bip->bli_format.blf_size = nvecs;
return;
}
/*
* Fill in an iovec for each set of contiguous chunks.
*/
first_bit = xfs_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size, 0);
ASSERT(first_bit != -1);
last_bit = first_bit;
nbits = 1;
for (;;) {
/*
* This takes the bit number to start looking from and
* returns the next set bit from there. It returns -1
* if there are no more bits set or the start bit is
* beyond the end of the bitmap.
*/
next_bit = xfs_next_bit(bip->bli_format.blf_data_map,
bip->bli_format.blf_map_size,
(uint)last_bit + 1);
/*
* If we run out of bits fill in the last iovec and get
* out of the loop.
* Else if we start a new set of bits then fill in the
* iovec for the series we were looking at and start
* counting the bits in the new one.
* Else we're still in the same set of bits so just
* keep counting and scanning.
*/
if (next_bit == -1) {
buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
break;
} else if (next_bit != last_bit + 1) {
buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
} else if (xfs_buf_offset(bp, next_bit << XFS_BLF_SHIFT) !=
(xfs_buf_offset(bp, last_bit << XFS_BLF_SHIFT) +
XFS_BLF_CHUNK)) {
buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
vecp->i_len = nbits * XFS_BLF_CHUNK;
vecp->i_type = XLOG_REG_TYPE_BCHUNK;
/* You would think we need to bump the nvecs here too, but we do not
* this number is used by recovery, and it gets confused by the boundary
* split here
* nvecs++;
*/
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
} else {
last_bit++;
nbits++;
}
}
bip->bli_format.blf_size = nvecs;
/*
* Check to make sure everything is consistent.
*/
trace_xfs_buf_item_format(bip);
xfs_buf_item_log_check(bip);
}
/*
* Writes a modified inode's changes out to the inode's on disk home.
* Originally based on xfs_iflush_int() from xfs_inode.c in the kernel.
*/
int
libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
{
xfs_inode_log_item_t *iip;
xfs_dinode_t *dip;
xfs_mount_t *mp;
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
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;
/* set *dip = inode's place in the buffer */
dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_boffset);
#ifdef DEBUG
ASSERT(ip->i_d.di_magic == XFS_DINODE_MAGIC);
if ((ip->i_d.di_mode & IFMT) == IFREG) {
ASSERT( (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS) ||
(ip->i_d.di_format == XFS_DINODE_FMT_BTREE) );
}
else if ((ip->i_d.di_mode & IFMT) == IFDIR) {
ASSERT( (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS) ||
(ip->i_d.di_format == XFS_DINODE_FMT_BTREE) ||
(ip->i_d.di_format == XFS_DINODE_FMT_LOCAL) );
}
ASSERT(ip->i_d.di_nextents+ip->i_d.di_anextents <= ip->i_d.di_nblocks);
ASSERT(ip->i_d.di_forkoff <= mp->m_sb.sb_inodesize);
#endif
/*
* Copy the dirty parts of the inode into the on-disk
* inode. We always copy out the core of the inode,
* because if the inode is dirty at all the core must
* be.
*/
xfs_xlate_dinode_core((xfs_caddr_t)&(dip->di_core), &(ip->i_d), -1,
ARCH_CONVERT);
/*
* If this is really an old format inode and the superblock version
* has not been updated to support only new format inodes, then
* 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) {
if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) {
/*
* Convert it back.
*/
ASSERT(ip->i_d.di_nlink <= XFS_MAXLINK_1);
INT_SET(dip->di_core.di_onlink, ARCH_CONVERT,
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;
INT_SET(dip->di_core.di_version, ARCH_CONVERT,
XFS_DINODE_VERSION_2);
ip->i_d.di_onlink = 0;
INT_ZERO(dip->di_core.di_onlink, ARCH_CONVERT);
bzero(&(ip->i_d.di_pad[0]), sizeof(ip->i_d.di_pad));
bzero(&(dip->di_core.di_pad[0]),
sizeof(dip->di_core.di_pad));
ASSERT(ip->i_d.di_projid == 0);
}
}
if (xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK, bp) == EFSCORRUPTED)
return EFSCORRUPTED;
if (XFS_IFORK_Q(ip)) {
/* The only error from xfs_iflush_fork is on the data fork. */
xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK, bp);
}
return 0;
}
