/*
* Validate a given inode number.
*/
int
xfs_dir_ino_validate(
xfs_mount_t *mp,
xfs_ino_t ino)
{
xfs_agblock_t agblkno;
xfs_agino_t agino;
xfs_agnumber_t agno;
int ino_ok;
int ioff;
agno = XFS_INO_TO_AGNO(mp, ino);
agblkno = XFS_INO_TO_AGBNO(mp, ino);
ioff = XFS_INO_TO_OFFSET(mp, ino);
agino = XFS_OFFBNO_TO_AGINO(mp, agblkno, ioff);
ino_ok =
agno < mp->m_sb.sb_agcount &&
agblkno < mp->m_sb.sb_agblocks &&
agblkno != 0 &&
ioff < (1 << mp->m_sb.sb_inopblog) &&
XFS_AGINO_TO_INO(mp, agno, agino) == ino;
if (unlikely(XFS_TEST_ERROR(!ino_ok, mp, XFS_ERRTAG_DIR_INO_VALIDATE,
XFS_RANDOM_DIR_INO_VALIDATE))) {
xfs_warn(mp, "Invalid inode number 0x%Lx",
(unsigned long long) ino);
XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
return XFS_ERROR(EFSCORRUPTED);
}
return 0;
}
/*
* Check that an inode's allocation status matches ir_free in the inobt
* record. First we try querying the in-core inode state, and if the inode
* isn't loaded we examine the on-disk inode directly.
*
* Since there can be 1:M and M:1 mappings between inobt records and inode
* clusters, we pass in the inode location information as an inobt record;
* the index of an inode cluster within the inobt record (as well as the
* cluster buffer itself); and the index of the inode within the cluster.
*
* @irec is the inobt record.
* @irec_ino is the inode offset from the start of the record.
* @dip is the on-disk inode.
*/
STATIC int
xchk_iallocbt_check_cluster_ifree(
struct xchk_btree *bs,
struct xfs_inobt_rec_incore *irec,
unsigned int irec_ino,
struct xfs_dinode *dip)
{
struct xfs_mount *mp = bs->cur->bc_mp;
xfs_ino_t fsino;
xfs_agino_t agino;
bool irec_free;
bool ino_inuse;
bool freemask_ok;
int error = 0;
if (xchk_should_terminate(bs->sc, &error))
return error;
/*
* Given an inobt record and the offset of an inode from the start of
* the record, compute which fs inode we're talking about.
*/
agino = irec->ir_startino + irec_ino;
fsino = XFS_AGINO_TO_INO(mp, bs->cur->bc_private.a.agno, agino);
irec_free = (irec->ir_free & XFS_INOBT_MASK(irec_ino));
if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC ||
(dip->di_version >= 3 && be64_to_cpu(dip->di_ino) != fsino)) {
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
goto out;
}
error = xfs_icache_inode_is_allocated(mp, bs->cur->bc_tp, fsino,
&ino_inuse);
if (error == -ENODATA) {
/* Not cached, just read the disk buffer */
freemask_ok = irec_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 = irec_free ^ ino_inuse;
}
if (!freemask_ok)
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
out:
return 0;
}
static void
process_inode(
xfs_agf_t *agf,
xfs_agino_t agino,
xfs_dinode_t *dip)
{
__uint64_t actual;
__uint64_t ideal;
xfs_ino_t ino;
int skipa;
int skipd;
ino = XFS_AGINO_TO_INO(mp, be32_to_cpu(agf->agf_seqno), agino);
switch (be16_to_cpu(dip->di_mode) & S_IFMT) {
case S_IFDIR:
skipd = !dflag;
break;
case S_IFREG:
if (!rflag && (be16_to_cpu(dip->di_flags) & XFS_DIFLAG_REALTIME))
skipd = 1;
else if (!Rflag &&
(ino == mp->m_sb.sb_rbmino ||
ino == mp->m_sb.sb_rsumino))
skipd = 1;
else if (!qflag &&
(ino == mp->m_sb.sb_uquotino ||
ino == mp->m_sb.sb_gquotino ||
ino == mp->m_sb.sb_pquotino))
skipd = 1;
else
skipd = !fflag;
break;
case S_IFLNK:
skipd = !lflag;
break;
default:
skipd = 1;
break;
}
actual = extcount_actual;
ideal = extcount_ideal;
if (!skipd)
process_fork(dip, XFS_DATA_FORK);
skipa = !aflag || !XFS_DFORK_Q(dip);
if (!skipa)
process_fork(dip, XFS_ATTR_FORK);
if (vflag && (!skipd || !skipa))
dbprintf(_("inode %lld actual %lld ideal %lld\n"),
ino, extcount_actual - actual, extcount_ideal - ideal);
}
/*
* Verify that an FS inode number pointer neither points outside the
* filesystem nor points at static AG metadata.
*/
bool
xfs_verify_ino(
struct xfs_mount *mp,
xfs_ino_t ino)
{
xfs_agnumber_t agno = XFS_INO_TO_AGNO(mp, ino);
xfs_agino_t agino = XFS_INO_TO_AGINO(mp, ino);
if (agno >= mp->m_sb.sb_agcount)
return false;
if (XFS_AGINO_TO_INO(mp, agno, agino) != ino)
return false;
return xfs_verify_agino(mp, agno, agino);
}
/*
* We are now using libxfs for our IO backend, so we should always try to use
* inode cluster buffers rather than filesystem block sized buffers for reading
* inodes. This means that we always use the same buffers as libxfs operations
* does, and that avoids buffer cache issues caused by overlapping buffers. This
* can be seen clearly when trying to read the root inode. Much of this logic is
* similar to libxfs_imap().
*/
void
set_cur_inode(
xfs_ino_t ino)
{
xfs_agblock_t agbno;
xfs_agino_t agino;
xfs_agnumber_t agno;
xfs_dinode_t *dip;
int offset;
int numblks = blkbb;
xfs_agblock_t cluster_agbno;
agno = XFS_INO_TO_AGNO(mp, ino);
agino = XFS_INO_TO_AGINO(mp, ino);
agbno = XFS_AGINO_TO_AGBNO(mp, agino);
offset = XFS_AGINO_TO_OFFSET(mp, agino);
if (agno >= mp->m_sb.sb_agcount || agbno >= mp->m_sb.sb_agblocks ||
offset >= mp->m_sb.sb_inopblock ||
XFS_AGINO_TO_INO(mp, agno, agino) != ino) {
dbprintf(_("bad inode number %lld\n"), ino);
return;
}
cur_agno = agno;
if (mp->m_inode_cluster_size > mp->m_sb.sb_blocksize &&
mp->m_inoalign_mask) {
xfs_agblock_t chunk_agbno;
xfs_agblock_t offset_agbno;
int blks_per_cluster;
blks_per_cluster = mp->m_inode_cluster_size >>
mp->m_sb.sb_blocklog;
offset_agbno = agbno & mp->m_inoalign_mask;
chunk_agbno = agbno - offset_agbno;
cluster_agbno = chunk_agbno +
((offset_agbno / blks_per_cluster) * blks_per_cluster);
offset += ((agbno - cluster_agbno) * mp->m_sb.sb_inopblock);
numblks = XFS_FSB_TO_BB(mp, blks_per_cluster);
} else
xfs_ino_t
xfs_agino_to_ino(xfs_mount_t *mp, xfs_agnumber_t a, xfs_agino_t i)
{
return XFS_AGINO_TO_INO(mp, a, i);
}
/*
* Allocate an inode on disk.
* Mode is used to tell whether the new inode will need space, and whether
* it is a directory.
*
* The arguments IO_agbp and alloc_done are defined to work within
* the constraint of one allocation per transaction.
* xfs_dialloc() is designed to be called twice if it has to do an
* allocation to make more free inodes. On the first call,
* IO_agbp should be set to NULL. If an inode is available,
* i.e., xfs_dialloc() did not need to do an allocation, an inode
* number is returned. In this case, IO_agbp would be set to the
* current ag_buf and alloc_done set to false.
* If an allocation needed to be done, xfs_dialloc would return
* the current ag_buf in IO_agbp and set alloc_done to true.
* The caller should then commit the current transaction, allocate a new
* transaction, and call xfs_dialloc() again, passing in the previous
* value of IO_agbp. IO_agbp should be held across the transactions.
* Since the agbp is locked across the two calls, the second call is
* guaranteed to have a free inode available.
*
* Once we successfully pick an inode its number is returned and the
* on-disk data structures are updated. The inode itself is not read
* in, since doing so would break ordering constraints with xfs_reclaim.
*/
int
xfs_dialloc(
xfs_trans_t *tp, /* transaction pointer */
xfs_ino_t parent, /* parent inode (directory) */
mode_t mode, /* mode bits for new inode */
int okalloc, /* ok to allocate more space */
xfs_buf_t **IO_agbp, /* in/out ag header's buffer */
boolean_t *alloc_done, /* true if we needed to replenish
inode freelist */
xfs_ino_t *inop) /* inode number allocated */
{
xfs_agnumber_t agcount; /* number of allocation groups */
xfs_buf_t *agbp; /* allocation group header's buffer */
xfs_agnumber_t agno; /* allocation group number */
xfs_agi_t *agi; /* allocation group header structure */
xfs_btree_cur_t *cur; /* inode allocation btree cursor */
int error; /* error return value */
int i; /* result code */
int ialloced; /* inode allocation status */
int noroom = 0; /* no space for inode blk allocation */
xfs_ino_t ino; /* fs-relative inode to be returned */
/* REFERENCED */
int j; /* result code */
xfs_mount_t *mp; /* file system mount structure */
int offset; /* index of inode in chunk */
xfs_agino_t pagino; /* parent's a.g. relative inode # */
xfs_agnumber_t pagno; /* parent's allocation group number */
xfs_inobt_rec_incore_t rec; /* inode allocation record */
xfs_agnumber_t tagno; /* testing allocation group number */
xfs_btree_cur_t *tcur; /* temp cursor */
xfs_inobt_rec_incore_t trec; /* temp inode allocation record */
if (*IO_agbp == NULL) {
/*
* We do not have an agbp, so select an initial allocation
* group for inode allocation.
*/
agbp = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
/*
* Couldn't find an allocation group satisfying the
* criteria, give up.
*/
if (!agbp) {
*inop = NULLFSINO;
return 0;
}
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
} else {
/*
* Continue where we left off before. In this case, we
* know that the allocation group has free inodes.
*/
agbp = *IO_agbp;
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
}
mp = tp->t_mountp;
agcount = mp->m_sb.sb_agcount;
agno = be32_to_cpu(agi->agi_seqno);
tagno = agno;
pagno = XFS_INO_TO_AGNO(mp, parent);
pagino = XFS_INO_TO_AGINO(mp, parent);
/*
* If we have already hit the ceiling of inode blocks then clear
* okalloc so we scan all available agi structures for a free
* inode.
*/
if (mp->m_maxicount &&
mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
noroom = 1;
okalloc = 0;
}
/*
* Loop until we find an allocation group that either has free inodes
* or in which we can allocate some inodes. Iterate through the
* allocation groups upward, wrapping at the end.
*/
*alloc_done = B_FALSE;
while (!agi->agi_freecount) {
/*
* Don't do anything if we're not supposed to allocate
* any blocks, just go on to the next ag.
*/
if (okalloc) {
/*
* Try to allocate some new inodes in the allocation
* group.
*/
if ((error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced))) {
xfs_trans_brelse(tp, agbp);
if (error == ENOSPC) {
*inop = NULLFSINO;
return 0;
} else
return error;
}
if (ialloced) {
/*
* We successfully allocated some inodes, return
* the current context to the caller so that it
* can commit the current transaction and call
* us again where we left off.
*/
ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
*alloc_done = B_TRUE;
*IO_agbp = agbp;
*inop = NULLFSINO;
return 0;
}
}
/*
* If it failed, give up on this ag.
*/
xfs_trans_brelse(tp, agbp);
/*
* Go on to the next ag: get its ag header.
*/
nextag:
if (++tagno == agcount)
tagno = 0;
if (tagno == agno) {
*inop = NULLFSINO;
return noroom ? ENOSPC : 0;
}
down_read(&mp->m_peraglock);
if (mp->m_perag[tagno].pagi_inodeok == 0) {
up_read(&mp->m_peraglock);
goto nextag;
}
error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
up_read(&mp->m_peraglock);
if (error)
goto nextag;
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC);
}
/*
* Here with an allocation group that has a free inode.
* Reset agno since we may have chosen a new ag in the
* loop above.
*/
agno = tagno;
*IO_agbp = NULL;
cur = xfs_btree_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno),
XFS_BTNUM_INO, (xfs_inode_t *)0, 0);
/*
* If pagino is 0 (this is the root inode allocation) use newino.
* This must work because we've just allocated some.
*/
if (!pagino)
pagino = be32_to_cpu(agi->agi_newino);
#ifdef DEBUG
if (cur->bc_nlevels == 1) {
int freecount = 0;
if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
do {
if ((error = xfs_inobt_get_rec(cur, &rec.ir_startino,
&rec.ir_freecount, &rec.ir_free, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
freecount += rec.ir_freecount;
if ((error = xfs_inobt_increment(cur, 0, &i)))
goto error0;
} while (i == 1);
ASSERT(freecount == be32_to_cpu(agi->agi_freecount) ||
XFS_FORCED_SHUTDOWN(mp));
}
#endif
/*
* If in the same a.g. as the parent, try to get near the parent.
*/
if (pagno == agno) {
if ((error = xfs_inobt_lookup_le(cur, pagino, 0, 0, &i)))
goto error0;
if (i != 0 &&
(error = xfs_inobt_get_rec(cur, &rec.ir_startino,
&rec.ir_freecount, &rec.ir_free, &j)) == 0 &&
j == 1 &&
rec.ir_freecount > 0) {
/*
* Found a free inode in the same chunk
* as parent, done.
*/
}
/*
* In the same a.g. as parent, but parent's chunk is full.
*/
else {
int doneleft; /* done, to the left */
int doneright; /* done, to the right */
if (error)
goto error0;
ASSERT(i == 1);
ASSERT(j == 1);
/*
* Duplicate the cursor, search left & right
* simultaneously.
*/
if ((error = xfs_btree_dup_cursor(cur, &tcur)))
goto error0;
/*
* Search left with tcur, back up 1 record.
*/
if ((error = xfs_inobt_decrement(tcur, 0, &i)))
goto error1;
doneleft = !i;
if (!doneleft) {
if ((error = xfs_inobt_get_rec(tcur,
&trec.ir_startino,
&trec.ir_freecount,
&trec.ir_free, &i)))
goto error1;
XFS_WANT_CORRUPTED_GOTO(i == 1, error1);
}
/*
* Search right with cur, go forward 1 record.
*/
if ((error = xfs_inobt_increment(cur, 0, &i)))
goto error1;
doneright = !i;
if (!doneright) {
if ((error = xfs_inobt_get_rec(cur,
&rec.ir_startino,
&rec.ir_freecount,
&rec.ir_free, &i)))
goto error1;
XFS_WANT_CORRUPTED_GOTO(i == 1, error1);
}
/*
* Loop until we find the closest inode chunk
* with a free one.
*/
while (!doneleft || !doneright) {
int useleft; /* using left inode
chunk this time */
/*
* Figure out which block is closer,
* if both are valid.
*/
if (!doneleft && !doneright)
useleft =
pagino -
(trec.ir_startino +
XFS_INODES_PER_CHUNK - 1) <
rec.ir_startino - pagino;
else
useleft = !doneleft;
/*
* If checking the left, does it have
* free inodes?
*/
if (useleft && trec.ir_freecount) {
/*
* Yes, set it up as the chunk to use.
*/
rec = trec;
xfs_btree_del_cursor(cur,
XFS_BTREE_NOERROR);
cur = tcur;
break;
}
/*
* If checking the right, does it have
* free inodes?
*/
if (!useleft && rec.ir_freecount) {
/*
* Yes, it's already set up.
*/
xfs_btree_del_cursor(tcur,
XFS_BTREE_NOERROR);
break;
}
/*
* If used the left, get another one
* further left.
*/
if (useleft) {
if ((error = xfs_inobt_decrement(tcur, 0,
&i)))
goto error1;
doneleft = !i;
if (!doneleft) {
if ((error = xfs_inobt_get_rec(
tcur,
&trec.ir_startino,
&trec.ir_freecount,
&trec.ir_free, &i)))
goto error1;
XFS_WANT_CORRUPTED_GOTO(i == 1,
error1);
}
}
/*
* If used the right, get another one
* further right.
*/
else {
if ((error = xfs_inobt_increment(cur, 0,
&i)))
goto error1;
doneright = !i;
if (!doneright) {
if ((error = xfs_inobt_get_rec(
cur,
&rec.ir_startino,
&rec.ir_freecount,
&rec.ir_free, &i)))
goto error1;
XFS_WANT_CORRUPTED_GOTO(i == 1,
error1);
}
}
}
ASSERT(!doneleft || !doneright);
}
}
/*
* In a different a.g. from the parent.
* See if the most recently allocated block has any free.
*/
else if (be32_to_cpu(agi->agi_newino) != NULLAGINO) {
if ((error = xfs_inobt_lookup_eq(cur,
be32_to_cpu(agi->agi_newino), 0, 0, &i)))
goto error0;
if (i == 1 &&
(error = xfs_inobt_get_rec(cur, &rec.ir_startino,
&rec.ir_freecount, &rec.ir_free, &j)) == 0 &&
j == 1 &&
rec.ir_freecount > 0) {
/*
* The last chunk allocated in the group still has
* a free inode.
*/
}
/*
* None left in the last group, search the whole a.g.
*/
else {
if (error)
goto error0;
if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
goto error0;
ASSERT(i == 1);
for (;;) {
if ((error = xfs_inobt_get_rec(cur,
&rec.ir_startino,
&rec.ir_freecount, &rec.ir_free,
&i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (rec.ir_freecount > 0)
break;
if ((error = xfs_inobt_increment(cur, 0, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
}
}
}
offset = XFS_IALLOC_FIND_FREE(&rec.ir_free);
ASSERT(offset >= 0);
ASSERT(offset < XFS_INODES_PER_CHUNK);
ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
XFS_INODES_PER_CHUNK) == 0);
ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
XFS_INOBT_CLR_FREE(&rec, offset);
rec.ir_freecount--;
if ((error = xfs_inobt_update(cur, rec.ir_startino, rec.ir_freecount,
rec.ir_free)))
goto error0;
be32_add_cpu(&agi->agi_freecount, -1);
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
down_read(&mp->m_peraglock);
mp->m_perag[tagno].pagi_freecount--;
up_read(&mp->m_peraglock);
#ifdef DEBUG
if (cur->bc_nlevels == 1) {
int freecount = 0;
if ((error = xfs_inobt_lookup_ge(cur, 0, 0, 0, &i)))
goto error0;
do {
if ((error = xfs_inobt_get_rec(cur, &rec.ir_startino,
&rec.ir_freecount, &rec.ir_free, &i)))
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
freecount += rec.ir_freecount;
if ((error = xfs_inobt_increment(cur, 0, &i)))
goto error0;
} while (i == 1);
ASSERT(freecount == be32_to_cpu(agi->agi_freecount) ||
XFS_FORCED_SHUTDOWN(mp));
}
#endif
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
*inop = ino;
return 0;
error1:
xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
return error;
}
/*
* Allocate an inode on disk.
* Mode is used to tell whether the new inode will need space, and whether
* it is a directory.
*
* The arguments IO_agbp and alloc_done are defined to work within
* the constraint of one allocation per transaction.
* xfs_dialloc() is designed to be called twice if it has to do an
* allocation to make more free inodes. On the first call,
* IO_agbp should be set to NULL. If an inode is available,
* i.e., xfs_dialloc() did not need to do an allocation, an inode
* number is returned. In this case, IO_agbp would be set to the
* current ag_buf and alloc_done set to false.
* If an allocation needed to be done, xfs_dialloc would return
* the current ag_buf in IO_agbp and set alloc_done to true.
* The caller should then commit the current transaction, allocate a new
* transaction, and call xfs_dialloc() again, passing in the previous
* value of IO_agbp. IO_agbp should be held across the transactions.
* Since the agbp is locked across the two calls, the second call is
* guaranteed to have a free inode available.
*
* Once we successfully pick an inode its number is returned and the
* on-disk data structures are updated. The inode itself is not read
* in, since doing so would break ordering constraints with xfs_reclaim.
*/
int
xfs_dialloc(
xfs_trans_t *tp, /* transaction pointer */
xfs_ino_t parent, /* parent inode (directory) */
umode_t mode, /* mode bits for new inode */
int okalloc, /* ok to allocate more space */
xfs_buf_t **IO_agbp, /* in/out ag header's buffer */
boolean_t *alloc_done, /* true if we needed to replenish
inode freelist */
xfs_ino_t *inop) /* inode number allocated */
{
xfs_agnumber_t agcount; /* number of allocation groups */
xfs_buf_t *agbp; /* allocation group header's buffer */
xfs_agnumber_t agno; /* allocation group number */
xfs_agi_t *agi; /* allocation group header structure */
xfs_btree_cur_t *cur; /* inode allocation btree cursor */
int error; /* error return value */
int i; /* result code */
int ialloced; /* inode allocation status */
int noroom = 0; /* no space for inode blk allocation */
xfs_ino_t ino; /* fs-relative inode to be returned */
/* REFERENCED */
int j; /* result code */
xfs_mount_t *mp; /* file system mount structure */
int offset; /* index of inode in chunk */
xfs_agino_t pagino; /* parent's AG relative inode # */
xfs_agnumber_t pagno; /* parent's AG number */
xfs_inobt_rec_incore_t rec; /* inode allocation record */
xfs_agnumber_t tagno; /* testing allocation group number */
xfs_btree_cur_t *tcur; /* temp cursor */
xfs_inobt_rec_incore_t trec; /* temp inode allocation record */
struct xfs_perag *pag;
if (*IO_agbp == NULL) {
/*
* We do not have an agbp, so select an initial allocation
* group for inode allocation.
*/
agbp = xfs_ialloc_ag_select(tp, parent, mode, okalloc);
/*
* Couldn't find an allocation group satisfying the
* criteria, give up.
*/
if (!agbp) {
*inop = NULLFSINO;
return 0;
}
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
} else {
/*
* Continue where we left off before. In this case, we
* know that the allocation group has free inodes.
*/
agbp = *IO_agbp;
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
}
mp = tp->t_mountp;
agcount = mp->m_sb.sb_agcount;
agno = be32_to_cpu(agi->agi_seqno);
tagno = agno;
pagno = XFS_INO_TO_AGNO(mp, parent);
pagino = XFS_INO_TO_AGINO(mp, parent);
/*
* If we have already hit the ceiling of inode blocks then clear
* okalloc so we scan all available agi structures for a free
* inode.
*/
if (mp->m_maxicount &&
mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
noroom = 1;
okalloc = 0;
}
/*
* Loop until we find an allocation group that either has free inodes
* or in which we can allocate some inodes. Iterate through the
* allocation groups upward, wrapping at the end.
*/
*alloc_done = B_FALSE;
while (!agi->agi_freecount) {
/*
* Don't do anything if we're not supposed to allocate
* any blocks, just go on to the next ag.
*/
if (okalloc) {
/*
* Try to allocate some new inodes in the allocation
* group.
*/
if ((error = xfs_ialloc_ag_alloc(tp, agbp, &ialloced))) {
xfs_trans_brelse(tp, agbp);
if (error == ENOSPC) {
*inop = NULLFSINO;
return 0;
} else
return error;
}
if (ialloced) {
/*
* We successfully allocated some inodes, return
* the current context to the caller so that it
* can commit the current transaction and call
* us again where we left off.
*/
ASSERT(be32_to_cpu(agi->agi_freecount) > 0);
*alloc_done = B_TRUE;
*IO_agbp = agbp;
*inop = NULLFSINO;
return 0;
}
}
/*
* If it failed, give up on this ag.
*/
xfs_trans_brelse(tp, agbp);
/*
* Go on to the next ag: get its ag header.
*/
nextag:
if (++tagno == agcount)
tagno = 0;
if (tagno == agno) {
*inop = NULLFSINO;
return noroom ? ENOSPC : 0;
}
pag = xfs_perag_get(mp, tagno);
if (pag->pagi_inodeok == 0) {
xfs_perag_put(pag);
goto nextag;
}
error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
xfs_perag_put(pag);
if (error)
goto nextag;
agi = XFS_BUF_TO_AGI(agbp);
ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
}
/*
* Here with an allocation group that has a free inode.
* Reset agno since we may have chosen a new ag in the
* loop above.
*/
agno = tagno;
*IO_agbp = NULL;
pag = xfs_perag_get(mp, agno);
restart_pagno:
cur = xfs_inobt_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno));
/*
* If pagino is 0 (this is the root inode allocation) use newino.
* This must work because we've just allocated some.
*/
if (!pagino)
pagino = be32_to_cpu(agi->agi_newino);
error = xfs_check_agi_freecount(cur, agi);
if (error)
goto error0;
/*
* If in the same AG as the parent, try to get near the parent.
*/
if (pagno == agno) {
int doneleft; /* done, to the left */
int doneright; /* done, to the right */
int searchdistance = 10;
error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
error = xfs_inobt_get_rec(cur, &rec, &j);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (rec.ir_freecount > 0) {
/*
* Found a free inode in the same chunk
* as the parent, done.
*/
goto alloc_inode;
}
/*
* In the same AG as parent, but parent's chunk is full.
*/
/* duplicate the cursor, search left & right simultaneously */
error = xfs_btree_dup_cursor(cur, &tcur);
if (error)
goto error0;
/*
* Skip to last blocks looked up if same parent inode.
*/
if (pagino != NULLAGINO &&
pag->pagl_pagino == pagino &&
pag->pagl_leftrec != NULLAGINO &&
pag->pagl_rightrec != NULLAGINO) {
error = xfs_ialloc_get_rec(tcur, pag->pagl_leftrec,
&trec, &doneleft, 1);
if (error)
goto error1;
error = xfs_ialloc_get_rec(cur, pag->pagl_rightrec,
&rec, &doneright, 0);
if (error)
goto error1;
} else {
/* search left with tcur, back up 1 record */
error = xfs_ialloc_next_rec(tcur, &trec, &doneleft, 1);
if (error)
goto error1;
/* search right with cur, go forward 1 record. */
error = xfs_ialloc_next_rec(cur, &rec, &doneright, 0);
if (error)
goto error1;
}
/*
* Loop until we find an inode chunk with a free inode.
*/
while (!doneleft || !doneright) {
int useleft; /* using left inode chunk this time */
if (!--searchdistance) {
/*
* Not in range - save last search
* location and allocate a new inode
*/
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto newino;
}
/* figure out the closer block if both are valid. */
if (!doneleft && !doneright) {
useleft = pagino -
(trec.ir_startino + XFS_INODES_PER_CHUNK - 1) <
rec.ir_startino - pagino;
} else {
useleft = !doneleft;
}
/* free inodes to the left? */
if (useleft && trec.ir_freecount) {
rec = trec;
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
cur = tcur;
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto alloc_inode;
}
/* free inodes to the right? */
if (!useleft && rec.ir_freecount) {
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto alloc_inode;
}
/* get next record to check */
if (useleft) {
error = xfs_ialloc_next_rec(tcur, &trec,
&doneleft, 1);
} else {
error = xfs_ialloc_next_rec(cur, &rec,
&doneright, 0);
}
if (error)
goto error1;
}
/*
* We've reached the end of the btree. because
* we are only searching a small chunk of the
* btree each search, there is obviously free
* inodes closer to the parent inode than we
* are now. restart the search again.
*/
pag->pagl_pagino = NULLAGINO;
pag->pagl_leftrec = NULLAGINO;
pag->pagl_rightrec = NULLAGINO;
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
goto restart_pagno;
}
/*
* In a different AG from the parent.
* See if the most recently allocated block has any free.
*/
newino:
if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
XFS_LOOKUP_EQ, &i);
if (error)
goto error0;
if (i == 1) {
error = xfs_inobt_get_rec(cur, &rec, &j);
if (error)
goto error0;
if (j == 1 && rec.ir_freecount > 0) {
/*
* The last chunk allocated in the group
* still has a free inode.
*/
goto alloc_inode;
}
}
}
/*
* None left in the last group, search the whole AG
*/
error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
for (;;) {
error = xfs_inobt_get_rec(cur, &rec, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (rec.ir_freecount > 0)
break;
error = xfs_btree_increment(cur, 0, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
}
alloc_inode:
offset = xfs_ialloc_find_free(&rec.ir_free);
ASSERT(offset >= 0);
ASSERT(offset < XFS_INODES_PER_CHUNK);
ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
XFS_INODES_PER_CHUNK) == 0);
ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
rec.ir_free &= ~XFS_INOBT_MASK(offset);
rec.ir_freecount--;
error = xfs_inobt_update(cur, &rec);
if (error)
goto error0;
be32_add_cpu(&agi->agi_freecount, -1);
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
pag->pagi_freecount--;
error = xfs_check_agi_freecount(cur, agi);
if (error)
goto error0;
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
xfs_perag_put(pag);
*inop = ino;
return 0;
error1:
xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
xfs_perag_put(pag);
return error;
}
/*
* Return inode number table for the filesystem.
*/
int /* error status */
xfs_inumbers(
struct xfs_mount *mp,/* mount point for filesystem */
xfs_ino_t *lastino,/* last inode returned */
int *count,/* size of buffer/count returned */
void __user *ubuffer,/* buffer with inode descriptions */
inumbers_fmt_pf formatter)
{
xfs_agnumber_t agno = XFS_INO_TO_AGNO(mp, *lastino);
xfs_agino_t agino = XFS_INO_TO_AGINO(mp, *lastino);
struct xfs_btree_cur *cur = NULL;
struct xfs_buf *agbp = NULL;
struct xfs_inogrp *buffer;
int bcount;
int left = *count;
int bufidx = 0;
int error = 0;
*count = 0;
if (agno >= mp->m_sb.sb_agcount ||
*lastino != XFS_AGINO_TO_INO(mp, agno, agino))
return error;
bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer)));
buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP);
do {
struct xfs_inobt_rec_incore r;
int stat;
if (!agbp) {
error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
if (error)
break;
cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
XFS_BTNUM_INO);
error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE,
&stat);
if (error)
break;
if (!stat)
goto next_ag;
}
error = xfs_inobt_get_rec(cur, &r, &stat);
if (error)
break;
if (!stat)
goto next_ag;
agino = r.ir_startino + XFS_INODES_PER_CHUNK - 1;
buffer[bufidx].xi_startino =
XFS_AGINO_TO_INO(mp, agno, r.ir_startino);
buffer[bufidx].xi_alloccount = r.ir_count - r.ir_freecount;
buffer[bufidx].xi_allocmask = ~r.ir_free;
if (++bufidx == bcount) {
long written;
error = formatter(ubuffer, buffer, bufidx, &written);
if (error)
break;
ubuffer += written;
*count += bufidx;
bufidx = 0;
}
if (!--left)
break;
error = xfs_btree_increment(cur, 0, &stat);
if (error)
break;
if (stat)
continue;
next_ag:
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
cur = NULL;
xfs_buf_relse(agbp);
agbp = NULL;
agino = 0;
agno++;
} while (agno < mp->m_sb.sb_agcount);
if (!error) {
if (bufidx) {
long written;
error = formatter(ubuffer, buffer, bufidx, &written);
if (!error)
*count += bufidx;
}
*lastino = XFS_AGINO_TO_INO(mp, agno, agino);
}
kmem_free(buffer);
if (cur)
xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
XFS_BTREE_NOERROR));
if (agbp)
xfs_buf_relse(agbp);
return error;
}
/*
* Return stat information in bulk (by-inode) for the filesystem.
*/
int /* error status */
xfs_bulkstat(
xfs_mount_t *mp, /* mount point for filesystem */
xfs_ino_t *lastinop, /* last inode returned */
int *ubcountp, /* size of buffer/count returned */
bulkstat_one_pf formatter, /* func that'd fill a single buf */
size_t statstruct_size, /* sizeof struct filling */
char __user *ubuffer, /* buffer with inode stats */
int *done) /* 1 if there are more stats to get */
{
xfs_buf_t *agbp; /* agi header buffer */
xfs_agino_t agino; /* inode # in allocation group */
xfs_agnumber_t agno; /* allocation group number */
xfs_btree_cur_t *cur; /* btree cursor for ialloc btree */
size_t irbsize; /* size of irec buffer in bytes */
xfs_inobt_rec_incore_t *irbuf; /* start of irec buffer */
int nirbuf; /* size of irbuf */
int ubcount; /* size of user's buffer */
struct xfs_bulkstat_agichunk ac;
int error = 0;
/*
* Get the last inode value, see if there's nothing to do.
*/
agno = XFS_INO_TO_AGNO(mp, *lastinop);
agino = XFS_INO_TO_AGINO(mp, *lastinop);
if (agno >= mp->m_sb.sb_agcount ||
*lastinop != XFS_AGINO_TO_INO(mp, agno, agino)) {
*done = 1;
*ubcountp = 0;
return 0;
}
ubcount = *ubcountp; /* statstruct's */
ac.ac_ubuffer = &ubuffer;
ac.ac_ubleft = ubcount * statstruct_size; /* bytes */;
ac.ac_ubelem = 0;
*ubcountp = 0;
*done = 0;
irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4);
if (!irbuf)
return -ENOMEM;
nirbuf = irbsize / sizeof(*irbuf);
/*
* Loop over the allocation groups, starting from the last
* inode returned; 0 means start of the allocation group.
*/
while (agno < mp->m_sb.sb_agcount) {
struct xfs_inobt_rec_incore *irbp = irbuf;
struct xfs_inobt_rec_incore *irbufend = irbuf + nirbuf;
bool end_of_ag = false;
int icount = 0;
int stat;
error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
if (error)
break;
/*
* Allocate and initialize a btree cursor for ialloc btree.
*/
cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
XFS_BTNUM_INO);
if (agino > 0) {
/*
* In the middle of an allocation group, we need to get
* the remainder of the chunk we're in.
*/
struct xfs_inobt_rec_incore r;
error = xfs_bulkstat_grab_ichunk(cur, agino, &icount, &r);
if (error)
goto del_cursor;
if (icount) {
irbp->ir_startino = r.ir_startino;
irbp->ir_holemask = r.ir_holemask;
irbp->ir_count = r.ir_count;
irbp->ir_freecount = r.ir_freecount;
irbp->ir_free = r.ir_free;
irbp++;
}
/* Increment to the next record */
error = xfs_btree_increment(cur, 0, &stat);
} else {
/* Start of ag. Lookup the first inode chunk */
error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &stat);
}
if (error || stat == 0) {
end_of_ag = true;
goto del_cursor;
}
/*
* Loop through inode btree records in this ag,
* until we run out of inodes or space in the buffer.
*/
while (irbp < irbufend && icount < ubcount) {
struct xfs_inobt_rec_incore r;
error = xfs_inobt_get_rec(cur, &r, &stat);
if (error || stat == 0) {
end_of_ag = true;
goto del_cursor;
}
/*
* If this chunk has any allocated inodes, save it.
* Also start read-ahead now for this chunk.
*/
if (r.ir_freecount < r.ir_count) {
xfs_bulkstat_ichunk_ra(mp, agno, &r);
irbp->ir_startino = r.ir_startino;
irbp->ir_holemask = r.ir_holemask;
irbp->ir_count = r.ir_count;
irbp->ir_freecount = r.ir_freecount;
irbp->ir_free = r.ir_free;
irbp++;
icount += r.ir_count - r.ir_freecount;
}
error = xfs_btree_increment(cur, 0, &stat);
if (error || stat == 0) {
end_of_ag = true;
goto del_cursor;
}
cond_resched();
}
/*
* Drop the btree buffers and the agi buffer as we can't hold any
* of the locks these represent when calling iget. If there is a
* pending error, then we are done.
*/
del_cursor:
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
xfs_buf_relse(agbp);
if (error)
break;
/*
* Now format all the good inodes into the user's buffer. The
* call to xfs_bulkstat_ag_ichunk() sets up the agino pointer
* for the next loop iteration.
*/
irbufend = irbp;
for (irbp = irbuf;
irbp < irbufend && ac.ac_ubleft >= statstruct_size;
irbp++) {
error = xfs_bulkstat_ag_ichunk(mp, agno, irbp,
formatter, statstruct_size, &ac,
&agino);
if (error)
break;
cond_resched();
}
/*
* If we've run out of space or had a formatting error, we
* are now done
*/
if (ac.ac_ubleft < statstruct_size || error)
break;
if (end_of_ag) {
agno++;
agino = 0;
}
}
/*
* Done, we're either out of filesystem or space to put the data.
*/
kmem_free(irbuf);
*ubcountp = ac.ac_ubelem;
/*
* We found some inodes, so clear the error status and return them.
* The lastino pointer will point directly at the inode that triggered
* any error that occurred, so on the next call the error will be
* triggered again and propagated to userspace as there will be no
* formatted inodes in the buffer.
*/
if (ac.ac_ubelem)
error = 0;
/*
* If we ran out of filesystem, lastino will point off the end of
* the filesystem so the next call will return immediately.
*/
*lastinop = XFS_AGINO_TO_INO(mp, agno, agino);
if (agno >= mp->m_sb.sb_agcount)
*done = 1;
return error;
}
/*
* Process inodes in chunk with a pointer to a formatter function
* that will iget the inode and fill in the appropriate structure.
*/
static int
xfs_bulkstat_ag_ichunk(
struct xfs_mount *mp,
xfs_agnumber_t agno,
struct xfs_inobt_rec_incore *irbp,
bulkstat_one_pf formatter,
size_t statstruct_size,
struct xfs_bulkstat_agichunk *acp,
xfs_agino_t *last_agino)
{
char __user **ubufp = acp->ac_ubuffer;
int chunkidx;
int error = 0;
xfs_agino_t agino = irbp->ir_startino;
for (chunkidx = 0; chunkidx < XFS_INODES_PER_CHUNK;
chunkidx++, agino++) {
int fmterror;
int ubused;
/* inode won't fit in buffer, we are done */
if (acp->ac_ubleft < statstruct_size)
break;
/* Skip if this inode is free */
if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free)
continue;
/* Get the inode and fill in a single buffer */
ubused = statstruct_size;
error = formatter(mp, XFS_AGINO_TO_INO(mp, agno, agino),
*ubufp, acp->ac_ubleft, &ubused, &fmterror);
if (fmterror == BULKSTAT_RV_GIVEUP ||
(error && error != -ENOENT && error != -EINVAL)) {
acp->ac_ubleft = 0;
ASSERT(error);
break;
}
/* be careful not to leak error if at end of chunk */
if (fmterror == BULKSTAT_RV_NOTHING || error) {
error = 0;
continue;
}
*ubufp += ubused;
acp->ac_ubleft -= ubused;
acp->ac_ubelem++;
}
/*
* Post-update *last_agino. At this point, agino will always point one
* inode past the last inode we processed successfully. Hence we
* substract that inode when setting the *last_agino cursor so that we
* return the correct cookie to userspace. On the next bulkstat call,
* the inode under the lastino cookie will be skipped as we have already
* processed it here.
*/
*last_agino = agino - 1;
return error;
}
/*
* Allocate an inode.
*
* The caller selected an AG for us, and made sure that free inodes are
* available.
*/
STATIC int
xfs_dialloc_ag(
struct xfs_trans *tp,
struct xfs_buf *agbp,
xfs_ino_t parent,
xfs_ino_t *inop)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
xfs_agnumber_t agno = be32_to_cpu(agi->agi_seqno);
xfs_agnumber_t pagno = XFS_INO_TO_AGNO(mp, parent);
xfs_agino_t pagino = XFS_INO_TO_AGINO(mp, parent);
struct xfs_perag *pag;
struct xfs_btree_cur *cur, *tcur;
struct xfs_inobt_rec_incore rec, trec;
xfs_ino_t ino;
int error;
int offset;
int i, j;
pag = xfs_perag_get(mp, agno);
ASSERT(pag->pagi_init);
ASSERT(pag->pagi_inodeok);
ASSERT(pag->pagi_freecount > 0);
restart_pagno:
cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
/*
* If pagino is 0 (this is the root inode allocation) use newino.
* This must work because we've just allocated some.
*/
if (!pagino)
pagino = be32_to_cpu(agi->agi_newino);
error = xfs_check_agi_freecount(cur, agi);
if (error)
goto error0;
/*
* If in the same AG as the parent, try to get near the parent.
*/
if (pagno == agno) {
int doneleft; /* done, to the left */
int doneright; /* done, to the right */
int searchdistance = 10;
error = xfs_inobt_lookup(cur, pagino, XFS_LOOKUP_LE, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
error = xfs_inobt_get_rec(cur, &rec, &j);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (rec.ir_freecount > 0) {
/*
* Found a free inode in the same chunk
* as the parent, done.
*/
goto alloc_inode;
}
/*
* In the same AG as parent, but parent's chunk is full.
*/
/* duplicate the cursor, search left & right simultaneously */
error = xfs_btree_dup_cursor(cur, &tcur);
if (error)
goto error0;
/*
* Skip to last blocks looked up if same parent inode.
*/
if (pagino != NULLAGINO &&
pag->pagl_pagino == pagino &&
pag->pagl_leftrec != NULLAGINO &&
pag->pagl_rightrec != NULLAGINO) {
error = xfs_ialloc_get_rec(tcur, pag->pagl_leftrec,
&trec, &doneleft);
if (error)
goto error1;
error = xfs_ialloc_get_rec(cur, pag->pagl_rightrec,
&rec, &doneright);
if (error)
goto error1;
} else {
/* search left with tcur, back up 1 record */
error = xfs_ialloc_next_rec(tcur, &trec, &doneleft, 1);
if (error)
goto error1;
/* search right with cur, go forward 1 record. */
error = xfs_ialloc_next_rec(cur, &rec, &doneright, 0);
if (error)
goto error1;
}
/*
* Loop until we find an inode chunk with a free inode.
*/
while (!doneleft || !doneright) {
int useleft; /* using left inode chunk this time */
if (!--searchdistance) {
/*
* Not in range - save last search
* location and allocate a new inode
*/
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto newino;
}
/* figure out the closer block if both are valid. */
if (!doneleft && !doneright) {
useleft = pagino -
(trec.ir_startino + XFS_INODES_PER_CHUNK - 1) <
rec.ir_startino - pagino;
} else {
useleft = !doneleft;
}
/* free inodes to the left? */
if (useleft && trec.ir_freecount) {
rec = trec;
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
cur = tcur;
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto alloc_inode;
}
/* free inodes to the right? */
if (!useleft && rec.ir_freecount) {
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
goto alloc_inode;
}
/* get next record to check */
if (useleft) {
error = xfs_ialloc_next_rec(tcur, &trec,
&doneleft, 1);
} else {
error = xfs_ialloc_next_rec(cur, &rec,
&doneright, 0);
}
if (error)
goto error1;
}
/*
* We've reached the end of the btree. because
* we are only searching a small chunk of the
* btree each search, there is obviously free
* inodes closer to the parent inode than we
* are now. restart the search again.
*/
pag->pagl_pagino = NULLAGINO;
pag->pagl_leftrec = NULLAGINO;
pag->pagl_rightrec = NULLAGINO;
xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
goto restart_pagno;
}
/*
* In a different AG from the parent.
* See if the most recently allocated block has any free.
*/
newino:
if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
XFS_LOOKUP_EQ, &i);
if (error)
goto error0;
if (i == 1) {
error = xfs_inobt_get_rec(cur, &rec, &j);
if (error)
goto error0;
if (j == 1 && rec.ir_freecount > 0) {
/*
* The last chunk allocated in the group
* still has a free inode.
*/
goto alloc_inode;
}
}
}
/*
* None left in the last group, search the whole AG
*/
error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
for (;;) {
error = xfs_inobt_get_rec(cur, &rec, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (rec.ir_freecount > 0)
break;
error = xfs_btree_increment(cur, 0, &i);
if (error)
goto error0;
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
}
alloc_inode:
offset = xfs_lowbit64(rec.ir_free);
ASSERT(offset >= 0);
ASSERT(offset < XFS_INODES_PER_CHUNK);
ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
XFS_INODES_PER_CHUNK) == 0);
ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
rec.ir_free &= ~XFS_INOBT_MASK(offset);
rec.ir_freecount--;
error = xfs_inobt_update(cur, &rec);
if (error)
goto error0;
be32_add_cpu(&agi->agi_freecount, -1);
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
pag->pagi_freecount--;
error = xfs_check_agi_freecount(cur, agi);
if (error)
goto error0;
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
xfs_perag_put(pag);
*inop = ino;
return 0;
error1:
xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
xfs_perag_put(pag);
return error;
}
/*
* Initialise a new set of inodes. When called without a transaction context
* (e.g. from recovery) we initiate a delayed write of the inode buffers rather
* than logging them (which in a transaction context puts them into the AIL
* for writeback rather than the xfsbufd queue).
*/
int
xfs_ialloc_inode_init(
struct xfs_mount *mp,
struct xfs_trans *tp,
struct list_head *buffer_list,
xfs_agnumber_t agno,
xfs_agblock_t agbno,
xfs_agblock_t length,
unsigned int gen)
{
struct xfs_buf *fbuf;
struct xfs_dinode *free;
int blks_per_cluster, nbufs, ninodes;
int version;
int i, j;
xfs_daddr_t d;
xfs_ino_t ino = 0;
/*
* Loop over the new block(s), filling in the inodes.
* For small block sizes, manipulate the inodes in buffers
* which are multiples of the blocks size.
*/
if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
blks_per_cluster = 1;
nbufs = length;
ninodes = mp->m_sb.sb_inopblock;
} else {
blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) /
mp->m_sb.sb_blocksize;
nbufs = length / blks_per_cluster;
ninodes = blks_per_cluster * mp->m_sb.sb_inopblock;
}
/*
* Figure out what version number to use in the inodes we create. If
* the superblock version has caught up to the one that supports the new
* inode format, then use the new inode version. Otherwise use the old
* version so that old kernels will continue to be able to use the file
* system.
*
* For v3 inodes, we also need to write the inode number into the inode,
* so calculate the first inode number of the chunk here as
* XFS_OFFBNO_TO_AGINO() only works within a filesystem block, not
* across multiple filesystem blocks (such as a cluster) and so cannot
* be used in the cluster buffer loop below.
*
* Further, because we are writing the inode directly into the buffer
* and calculating a CRC on the entire inode, we have ot log the entire
* inode so that the entire range the CRC covers is present in the log.
* That means for v3 inode we log the entire buffer rather than just the
* inode cores.
*/
if (xfs_sb_version_hascrc(&mp->m_sb)) {
version = 3;
ino = XFS_AGINO_TO_INO(mp, agno,
XFS_OFFBNO_TO_AGINO(mp, agbno, 0));
/*
* log the initialisation that is about to take place as an
* logical operation. This means the transaction does not
* need to log the physical changes to the inode buffers as log
* recovery will know what initialisation is actually needed.
* Hence we only need to log the buffers as "ordered" buffers so
* they track in the AIL as if they were physically logged.
*/
if (tp)
xfs_icreate_log(tp, agno, agbno, XFS_IALLOC_INODES(mp),
mp->m_sb.sb_inodesize, length, gen);
} else if (xfs_sb_version_hasnlink(&mp->m_sb))
version = 2;
else
version = 1;
for (j = 0; j < nbufs; j++) {
/*
* Get the block.
*/
d = XFS_AGB_TO_DADDR(mp, agno, agbno + (j * blks_per_cluster));
fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
mp->m_bsize * blks_per_cluster,
XBF_UNMAPPED);
if (!fbuf)
return ENOMEM;
/* Initialize the inode buffers and log them appropriately. */
fbuf->b_ops = &xfs_inode_buf_ops;
xfs_buf_zero(fbuf, 0, BBTOB(fbuf->b_length));
for (i = 0; i < ninodes; i++) {
int ioffset = i << mp->m_sb.sb_inodelog;
uint isize = xfs_dinode_size(version);
free = xfs_make_iptr(mp, fbuf, i);
free->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
free->di_version = version;
free->di_gen = cpu_to_be32(gen);
free->di_next_unlinked = cpu_to_be32(NULLAGINO);
if (version == 3) {
free->di_ino = cpu_to_be64(ino);
ino++;
uuid_copy(&free->di_uuid, &mp->m_sb.sb_uuid);
xfs_dinode_calc_crc(mp, free);
} else if (tp) {
/* just log the inode core */
xfs_trans_log_buf(tp, fbuf, ioffset,
ioffset + isize - 1);
}
}
if (tp) {
/*
* Mark the buffer as an inode allocation buffer so it
* sticks in AIL at the point of this allocation
* transaction. This ensures the they are on disk before
* the tail of the log can be moved past this
* transaction (i.e. by preventing relogging from moving
* it forward in the log).
*/
xfs_trans_inode_alloc_buf(tp, fbuf);
if (version == 3) {
/*
* Mark the buffer as ordered so that they are
* not physically logged in the transaction but
* still tracked in the AIL as part of the
* transaction and pin the log appropriately.
*/
xfs_trans_ordered_buf(tp, fbuf);
xfs_trans_log_buf(tp, fbuf, 0,
BBTOB(fbuf->b_length) - 1);
}
} else {
fbuf->b_flags |= XBF_DONE;
xfs_buf_delwri_queue(fbuf, buffer_list);
xfs_buf_relse(fbuf);
}
}
return 0;
}
/*
* Initialise a new set of inodes.
*/
STATIC int
xfs_ialloc_inode_init(
struct xfs_mount *mp,
struct xfs_trans *tp,
xfs_agnumber_t agno,
xfs_agblock_t agbno,
xfs_agblock_t length,
unsigned int gen)
{
struct xfs_buf *fbuf;
struct xfs_dinode *free;
int blks_per_cluster, nbufs, ninodes;
int version;
int i, j;
xfs_daddr_t d;
xfs_ino_t ino = 0;
/*
* Loop over the new block(s), filling in the inodes.
* For small block sizes, manipulate the inodes in buffers
* which are multiples of the blocks size.
*/
if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
blks_per_cluster = 1;
nbufs = length;
ninodes = mp->m_sb.sb_inopblock;
} else {
blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) /
mp->m_sb.sb_blocksize;
nbufs = length / blks_per_cluster;
ninodes = blks_per_cluster * mp->m_sb.sb_inopblock;
}
/*
* Figure out what version number to use in the inodes we create. If
* the superblock version has caught up to the one that supports the new
* inode format, then use the new inode version. Otherwise use the old
* version so that old kernels will continue to be able to use the file
* system.
*
* For v3 inodes, we also need to write the inode number into the inode,
* so calculate the first inode number of the chunk here as
* XFS_OFFBNO_TO_AGINO() only works within a filesystem block, not
* across multiple filesystem blocks (such as a cluster) and so cannot
* be used in the cluster buffer loop below.
*
* Further, because we are writing the inode directly into the buffer
* and calculating a CRC on the entire inode, we have ot log the entire
* inode so that the entire range the CRC covers is present in the log.
* That means for v3 inode we log the entire buffer rather than just the
* inode cores.
*/
if (xfs_sb_version_hascrc(&mp->m_sb)) {
version = 3;
ino = XFS_AGINO_TO_INO(mp, agno,
XFS_OFFBNO_TO_AGINO(mp, agbno, 0));
} else if (xfs_sb_version_hasnlink(&mp->m_sb))
version = 2;
else
version = 1;
for (j = 0; j < nbufs; j++) {
/*
* Get the block.
*/
d = XFS_AGB_TO_DADDR(mp, agno, agbno + (j * blks_per_cluster));
fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
mp->m_bsize * blks_per_cluster,
XBF_UNMAPPED);
if (!fbuf)
return ENOMEM;
/*
* Initialize all inodes in this buffer and then log them.
*
* XXX: It would be much better if we had just one transaction
* to log a whole cluster of inodes instead of all the
* individual transactions causing a lot of log traffic.
*/
fbuf->b_ops = &xfs_inode_buf_ops;
xfs_buf_zero(fbuf, 0, BBTOB(fbuf->b_length));
for (i = 0; i < ninodes; i++) {
int ioffset = i << mp->m_sb.sb_inodelog;
uint isize = xfs_dinode_size(version);
free = xfs_make_iptr(mp, fbuf, i);
free->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
free->di_version = version;
free->di_gen = cpu_to_be32(gen);
free->di_next_unlinked = cpu_to_be32(NULLAGINO);
if (version == 3) {
free->di_ino = cpu_to_be64(ino);
ino++;
uuid_copy(&free->di_uuid, &mp->m_sb.sb_uuid);
xfs_dinode_calc_crc(mp, free);
} else {
/* just log the inode core */
xfs_trans_log_buf(tp, fbuf, ioffset,
ioffset + isize - 1);
}
}
if (version == 3) {
/* need to log the entire buffer */
xfs_trans_log_buf(tp, fbuf, 0,
BBTOB(fbuf->b_length) - 1);
}
xfs_trans_inode_alloc_buf(tp, fbuf);
}
return 0;
}
static int
libxfs_initialize_perag(
xfs_mount_t *mp,
xfs_agnumber_t agcount,
xfs_agnumber_t *maxagi)
{
xfs_agnumber_t index, max_metadata;
xfs_agnumber_t first_initialised = 0;
xfs_perag_t *pag;
xfs_agino_t agino;
xfs_ino_t ino;
xfs_sb_t *sbp = &mp->m_sb;
int error = -ENOMEM;
/*
* Walk the current per-ag tree so we don't try to initialise AGs
* that already exist (growfs case). Allocate and insert all the
* AGs we don't find ready for initialisation.
*/
for (index = 0; index < agcount; index++) {
pag = xfs_perag_get(mp, index);
if (pag) {
xfs_perag_put(pag);
continue;
}
if (!first_initialised)
first_initialised = index;
pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL);
if (!pag)
goto out_unwind;
pag->pag_agno = index;
pag->pag_mount = mp;
if (radix_tree_insert(&mp->m_perag_tree, index, pag)) {
error = -EEXIST;
goto out_unwind;
}
}
/*
* If we mount with the inode64 option, or no inode overflows
* the legacy 32-bit address space clear the inode32 option.
*/
agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);
if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) && ino > XFS_MAXINUMBER_32)
mp->m_flags |= XFS_MOUNT_32BITINODES;
else
mp->m_flags &= ~XFS_MOUNT_32BITINODES;
if (mp->m_flags & XFS_MOUNT_32BITINODES) {
/*
* Calculate how much should be reserved for inodes to meet
* the max inode percentage.
*/
if (mp->m_maxicount) {
__uint64_t icount;
icount = sbp->sb_dblocks * sbp->sb_imax_pct;
do_div(icount, 100);
icount += sbp->sb_agblocks - 1;
do_div(icount, sbp->sb_agblocks);
max_metadata = icount;
} else {
max_metadata = agcount;
}
for (index = 0; index < agcount; index++) {
ino = XFS_AGINO_TO_INO(mp, index, agino);
if (ino > XFS_MAXINUMBER_32) {
index++;
break;
}
pag = xfs_perag_get(mp, index);
pag->pagi_inodeok = 1;
if (index < max_metadata)
pag->pagf_metadata = 1;
xfs_perag_put(pag);
}
} else {
for (index = 0; index < agcount; index++) {
pag = xfs_perag_get(mp, index);
pag->pagi_inodeok = 1;
xfs_perag_put(pag);
}
}
if (maxagi)
*maxagi = index;
return 0;
out_unwind:
kmem_free(pag);
for (; index > first_initialised; index--) {
pag = radix_tree_delete(&mp->m_perag_tree, index);
kmem_free(pag);
}
return error;
}
/* Make sure the free mask is consistent with what the inodes think. */
STATIC int
xfs_scrub_iallocbt_check_freemask(
struct xfs_scrub_btree *bs,
struct xfs_inobt_rec_incore *irec)
{
struct xfs_owner_info oinfo;
struct xfs_imap imap;
struct xfs_mount *mp = bs->cur->bc_mp;
struct xfs_dinode *dip;
struct xfs_buf *bp;
xfs_ino_t fsino;
xfs_agino_t nr_inodes;
xfs_agino_t agino;
xfs_agino_t chunkino;
xfs_agino_t clusterino;
xfs_agblock_t agbno;
int blks_per_cluster;
uint16_t holemask;
uint16_t ir_holemask;
int error = 0;
/* Make sure the freemask matches the inode records. */
blks_per_cluster = xfs_icluster_size_fsb(mp);
nr_inodes = XFS_OFFBNO_TO_AGINO(mp, blks_per_cluster, 0);
xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_INODES);
for (agino = irec->ir_startino;
agino < irec->ir_startino + XFS_INODES_PER_CHUNK;
agino += blks_per_cluster * mp->m_sb.sb_inopblock) {
fsino = XFS_AGINO_TO_INO(mp, bs->cur->bc_private.a.agno, agino);
chunkino = agino - irec->ir_startino;
agbno = XFS_AGINO_TO_AGBNO(mp, agino);
/* Compute the holemask mask for this cluster. */
for (clusterino = 0, holemask = 0; clusterino < nr_inodes;
clusterino += XFS_INODES_PER_HOLEMASK_BIT)
holemask |= XFS_INOBT_MASK((chunkino + clusterino) /
XFS_INODES_PER_HOLEMASK_BIT);
/* The whole cluster must be a hole or not a hole. */
ir_holemask = (irec->ir_holemask & holemask);
if (ir_holemask != holemask && ir_holemask != 0) {
xfs_scrub_btree_set_corrupt(bs->sc, bs->cur, 0);
continue;
}
/* If any part of this is a hole, skip it. */
if (ir_holemask)
continue;
/* Grab the inode cluster buffer. */
imap.im_blkno = XFS_AGB_TO_DADDR(mp, bs->cur->bc_private.a.agno,
agbno);
imap.im_len = XFS_FSB_TO_BB(mp, blks_per_cluster);
imap.im_boffset = 0;
error = xfs_imap_to_bp(mp, bs->cur->bc_tp, &imap,
&dip, &bp, 0, 0);
if (!xfs_scrub_btree_process_error(bs->sc, bs->cur, 0, &error))
continue;
/* Which inodes are free? */
for (clusterino = 0; clusterino < nr_inodes; clusterino++) {
error = xfs_scrub_iallocbt_check_cluster_freemask(bs,
fsino, chunkino, clusterino, irec, bp);
if (error) {
xfs_trans_brelse(bs->cur->bc_tp, bp);
return error;
}
}
xfs_trans_brelse(bs->cur->bc_tp, bp);
}
return error;
}
