int dlm_send_proxy_ast_msg(struct dlm_ctxt *dlm, struct dlm_lock_resource *res,
struct dlm_lock *lock, int msg_type,
int blocked_type, int flags)
{
int ret = 0;
struct dlm_proxy_ast past;
struct kvec vec[2];
size_t veclen = 1;
int status;
mlog(0, "%s: res %.*s, to %u, type %d, blocked_type %d\n", dlm->name,
res->lockname.len, res->lockname.name, lock->ml.node, msg_type,
blocked_type);
memset(&past, 0, sizeof(struct dlm_proxy_ast));
past.node_idx = dlm->node_num;
past.type = msg_type;
past.blocked_type = blocked_type;
past.namelen = res->lockname.len;
memcpy(past.name, res->lockname.name, past.namelen);
past.cookie = lock->ml.cookie;
vec[0].iov_len = sizeof(struct dlm_proxy_ast);
vec[0].iov_base = &past;
if (flags & DLM_LKSB_GET_LVB) {
be32_add_cpu(&past.flags, LKM_GET_LVB);
vec[1].iov_len = DLM_LVB_LEN;
vec[1].iov_base = lock->lksb->lvb;
veclen++;
}
ret = o2net_send_message_vec(DLM_PROXY_AST_MSG, dlm->key, vec, veclen,
lock->ml.node, &status);
if (ret < 0)
mlog(ML_ERROR, "%s: res %.*s, error %d send AST to node %u\n",
dlm->name, res->lockname.len, res->lockname.name, ret,
lock->ml.node);
else {
if (status == DLM_RECOVERING) {
mlog(ML_ERROR, "sent AST to node %u, it thinks this "
"node is dead!\n", lock->ml.node);
BUG();
} else if (status == DLM_MIGRATING) {
mlog(ML_ERROR, "sent AST to node %u, it returned "
"DLM_MIGRATING!\n", lock->ml.node);
BUG();
} else if (status != DLM_NORMAL && status != DLM_IVLOCKID) {
mlog(ML_ERROR, "AST to node %u returned %d!\n",
lock->ml.node, status);
/* ignore it */
}
ret = 0;
}
return ret;
}
STATIC void
xfs_inobt_set_root(
struct xfs_btree_cur *cur,
union xfs_btree_ptr *nptr,
int inc) /* level change */
{
struct xfs_buf *agbp = cur->bc_private.a.agbp;
struct xfs_agi *agi = XFS_BUF_TO_AGI(agbp);
agi->agi_root = nptr->s;
be32_add_cpu(&agi->agi_level, inc);
xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL);
}
/*
* compact the leaf entries.
* Leave the highest-numbered stale entry stale.
* XXX should be the one closest to mid but mid is not yet computed.
*/
static void
xfs_dir2_block_compact(
struct xfs_trans *tp,
struct xfs_inode *dp,
struct xfs_buf *bp,
struct xfs_dir2_data_hdr *hdr,
struct xfs_dir2_block_tail *btp,
struct xfs_dir2_leaf_entry *blp,
int *needlog,
int *lfloghigh,
int *lfloglow)
{
int fromidx; /* source leaf index */
int toidx; /* target leaf index */
int needscan = 0;
int highstale; /* high stale index */
fromidx = toidx = be32_to_cpu(btp->count) - 1;
highstale = *lfloghigh = -1;
for (; fromidx >= 0; fromidx--) {
if (blp[fromidx].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) {
if (highstale == -1)
highstale = toidx;
else {
if (*lfloghigh == -1)
*lfloghigh = toidx;
continue;
}
}
if (fromidx < toidx)
blp[toidx] = blp[fromidx];
toidx--;
}
*lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
*lfloghigh -= be32_to_cpu(btp->stale) - 1;
be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
xfs_dir2_data_make_free(tp, dp, bp,
(xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
(xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
needlog, &needscan);
blp += be32_to_cpu(btp->stale) - 1;
btp->stale = cpu_to_be32(1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
*/
if (needscan)
xfs_dir2_data_freescan(dp, hdr, needlog);
}
STATIC void
xfs_allocbt_set_root(
struct xfs_btree_cur *cur,
union xfs_btree_ptr *ptr,
int inc)
{
struct xfs_buf *agbp = cur->bc_private.a.agbp;
struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
int btnum = cur->bc_btnum;
ASSERT(ptr->s != 0);
agf->agf_roots[btnum] = ptr->s;
be32_add_cpu(&agf->agf_levels[btnum], inc);
cur->bc_mp->m_perag[seqno].pagf_levels[btnum] += inc;
xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
}
STATIC void
xfs_refcountbt_set_root(
struct xfs_btree_cur *cur,
union xfs_btree_ptr *ptr,
int inc)
{
struct xfs_buf *agbp = cur->bc_private.a.agbp;
struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
struct xfs_perag *pag = xfs_perag_get(cur->bc_mp, seqno);
ASSERT(ptr->s != 0);
agf->agf_refcount_root = ptr->s;
be32_add_cpu(&agf->agf_refcount_level, inc);
pag->pagf_refcount_level += inc;
xfs_perag_put(pag);
xfs_alloc_log_agf(cur->bc_tp, agbp,
XFS_AGF_REFCOUNT_ROOT | XFS_AGF_REFCOUNT_LEVEL);
}
static ssize_t o2nm_node_ipv4_address_write(struct o2nm_node *node,
const char *page,
size_t count)
{
struct o2nm_cluster *cluster = to_o2nm_cluster_from_node(node);
int ret, i;
struct rb_node **p, *parent;
unsigned int octets[4];
__be32 ipv4_addr = 0;
ret = sscanf(page, "%3u.%3u.%3u.%3u", &octets[3], &octets[2],
&octets[1], &octets[0]);
if (ret != 4)
return -EINVAL;
for (i = 0; i < ARRAY_SIZE(octets); i++) {
if (octets[i] > 255)
return -ERANGE;
be32_add_cpu(&ipv4_addr, octets[i] << (i * 8));
}
ret = 0;
write_lock(&cluster->cl_nodes_lock);
if (o2nm_node_ip_tree_lookup(cluster, ipv4_addr, &p, &parent))
ret = -EEXIST;
else {
rb_link_node(&node->nd_ip_node, parent, p);
rb_insert_color(&node->nd_ip_node, &cluster->cl_node_ip_tree);
}
write_unlock(&cluster->cl_nodes_lock);
if (ret)
return ret;
memcpy(&node->nd_ipv4_address, &ipv4_addr, sizeof(ipv4_addr));
return count;
}
/*
* Remove an entry from a block format directory.
* If that makes the block small enough to fit in shortform, transform it.
*/
int /* error */
xfs_dir2_block_removename(
xfs_da_args_t *args) /* directory operation args */
{
xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf pointer */
struct xfs_buf *bp; /* block buffer */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* incore inode */
int ent; /* block leaf entry index */
int error; /* error return value */
int needlog; /* need to log block header */
int needscan; /* need to fixup bestfree */
xfs_dir2_sf_hdr_t sfh; /* shortform header */
int size; /* shortform size */
xfs_trans_t *tp; /* transaction pointer */
trace_xfs_dir2_block_removename(args);
/*
* Look up the entry in the block. Gets the buffer and entry index.
* It will always be there, the vnodeops level does a lookup first.
*/
if ((error = xfs_dir2_block_lookup_int(args, &bp, &ent))) {
return error;
}
dp = args->dp;
tp = args->trans;
hdr = bp->b_addr;
btp = xfs_dir2_block_tail_p(args->geo, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Point to the data entry using the leaf entry.
*/
dep = (xfs_dir2_data_entry_t *)((char *)hdr +
xfs_dir2_dataptr_to_off(args->geo,
be32_to_cpu(blp[ent].address)));
/*
* Mark the data entry's space free.
*/
needlog = needscan = 0;
xfs_dir2_data_make_free(args, bp,
(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
/*
* Fix up the block tail.
*/
be32_add_cpu(&btp->stale, 1);
xfs_dir2_block_log_tail(tp, bp);
/*
* Remove the leaf entry by marking it stale.
*/
blp[ent].address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
xfs_dir2_block_log_leaf(tp, bp, ent, ent);
/*
* Fix up bestfree, log the header if necessary.
*/
if (needscan)
xfs_dir2_data_freescan(dp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(args, bp);
xfs_dir3_data_check(dp, bp);
/*
* See if the size as a shortform is good enough.
*/
size = xfs_dir2_block_sfsize(dp, hdr, &sfh);
if (size > XFS_IFORK_DSIZE(dp))
return 0;
/*
* If it works, do the conversion.
*/
return xfs_dir2_block_to_sf(args, bp, size, &sfh);
}
static int hfsplus_fill_super(struct super_block *sb, void *data, int silent)
{
struct hfsplus_vh *vhdr;
struct hfsplus_sb_info *sbi;
hfsplus_cat_entry entry;
struct hfs_find_data fd;
struct inode *root, *inode;
struct qstr str;
struct nls_table *nls = NULL;
int err;
err = -EINVAL;
sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
if (!sbi)
goto out;
sb->s_fs_info = sbi;
mutex_init(&sbi->alloc_mutex);
mutex_init(&sbi->vh_mutex);
hfsplus_fill_defaults(sbi);
err = -EINVAL;
if (!hfsplus_parse_options(data, sbi)) {
printk(KERN_ERR "hfs: unable to parse mount options\n");
goto out_unload_nls;
}
/* temporarily use utf8 to correctly find the hidden dir below */
nls = sbi->nls;
sbi->nls = load_nls("utf8");
if (!sbi->nls) {
printk(KERN_ERR "hfs: unable to load nls for utf8\n");
goto out_unload_nls;
}
/* Grab the volume header */
if (hfsplus_read_wrapper(sb)) {
if (!silent)
printk(KERN_WARNING "hfs: unable to find HFS+ superblock\n");
goto out_unload_nls;
}
vhdr = sbi->s_vhdr;
/* Copy parts of the volume header into the superblock */
sb->s_magic = HFSPLUS_VOLHEAD_SIG;
if (be16_to_cpu(vhdr->version) < HFSPLUS_MIN_VERSION ||
be16_to_cpu(vhdr->version) > HFSPLUS_CURRENT_VERSION) {
printk(KERN_ERR "hfs: wrong filesystem version\n");
goto out_free_vhdr;
}
sbi->total_blocks = be32_to_cpu(vhdr->total_blocks);
sbi->free_blocks = be32_to_cpu(vhdr->free_blocks);
sbi->next_cnid = be32_to_cpu(vhdr->next_cnid);
sbi->file_count = be32_to_cpu(vhdr->file_count);
sbi->folder_count = be32_to_cpu(vhdr->folder_count);
sbi->data_clump_blocks =
be32_to_cpu(vhdr->data_clump_sz) >> sbi->alloc_blksz_shift;
if (!sbi->data_clump_blocks)
sbi->data_clump_blocks = 1;
sbi->rsrc_clump_blocks =
be32_to_cpu(vhdr->rsrc_clump_sz) >> sbi->alloc_blksz_shift;
if (!sbi->rsrc_clump_blocks)
sbi->rsrc_clump_blocks = 1;
/* Set up operations so we can load metadata */
sb->s_op = &hfsplus_sops;
sb->s_maxbytes = MAX_LFS_FILESIZE;
if (!(vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_UNMNT))) {
printk(KERN_WARNING "hfs: Filesystem was "
"not cleanly unmounted, "
"running fsck.hfsplus is recommended. "
"mounting read-only.\n");
sb->s_flags |= MS_RDONLY;
} else if (test_and_clear_bit(HFSPLUS_SB_FORCE, &sbi->flags)) {
/* nothing */
} else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) {
printk(KERN_WARNING "hfs: Filesystem is marked locked, mounting read-only.\n");
sb->s_flags |= MS_RDONLY;
} else if ((vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) &&
!(sb->s_flags & MS_RDONLY)) {
printk(KERN_WARNING "hfs: write access to "
"a journaled filesystem is not supported, "
"use the force option at your own risk, "
"mounting read-only.\n");
sb->s_flags |= MS_RDONLY;
}
/* Load metadata objects (B*Trees) */
sbi->ext_tree = hfs_btree_open(sb, HFSPLUS_EXT_CNID);
if (!sbi->ext_tree) {
printk(KERN_ERR "hfs: failed to load extents file\n");
goto out_free_vhdr;
}
sbi->cat_tree = hfs_btree_open(sb, HFSPLUS_CAT_CNID);
if (!sbi->cat_tree) {
printk(KERN_ERR "hfs: failed to load catalog file\n");
goto out_close_ext_tree;
}
inode = hfsplus_iget(sb, HFSPLUS_ALLOC_CNID);
if (IS_ERR(inode)) {
printk(KERN_ERR "hfs: failed to load allocation file\n");
err = PTR_ERR(inode);
goto out_close_cat_tree;
}
sbi->alloc_file = inode;
/* Load the root directory */
root = hfsplus_iget(sb, HFSPLUS_ROOT_CNID);
if (IS_ERR(root)) {
printk(KERN_ERR "hfs: failed to load root directory\n");
err = PTR_ERR(root);
goto out_put_alloc_file;
}
str.len = sizeof(HFSP_HIDDENDIR_NAME) - 1;
str.name = HFSP_HIDDENDIR_NAME;
hfs_find_init(sbi->cat_tree, &fd);
hfsplus_cat_build_key(sb, fd.search_key, HFSPLUS_ROOT_CNID, &str);
if (!hfs_brec_read(&fd, &entry, sizeof(entry))) {
hfs_find_exit(&fd);
if (entry.type != cpu_to_be16(HFSPLUS_FOLDER))
goto out_put_root;
inode = hfsplus_iget(sb, be32_to_cpu(entry.folder.id));
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto out_put_root;
}
sbi->hidden_dir = inode;
} else
hfs_find_exit(&fd);
if (!(sb->s_flags & MS_RDONLY)) {
/*
* H+LX == hfsplusutils, H+Lx == this driver, H+lx is unused
* all three are registered with Apple for our use
*/
vhdr->last_mount_vers = cpu_to_be32(HFSP_MOUNT_VERSION);
vhdr->modify_date = hfsp_now2mt();
be32_add_cpu(&vhdr->write_count, 1);
vhdr->attributes &= cpu_to_be32(~HFSPLUS_VOL_UNMNT);
vhdr->attributes |= cpu_to_be32(HFSPLUS_VOL_INCNSTNT);
hfsplus_sync_fs(sb, 1);
if (!sbi->hidden_dir) {
mutex_lock(&sbi->vh_mutex);
sbi->hidden_dir = hfsplus_new_inode(sb, S_IFDIR);
hfsplus_create_cat(sbi->hidden_dir->i_ino, root, &str,
sbi->hidden_dir);
mutex_unlock(&sbi->vh_mutex);
hfsplus_mark_inode_dirty(sbi->hidden_dir,
HFSPLUS_I_CAT_DIRTY);
}
}
sb->s_d_op = &hfsplus_dentry_operations;
sb->s_root = d_alloc_root(root);
if (!sb->s_root) {
err = -ENOMEM;
goto out_put_hidden_dir;
}
unload_nls(sbi->nls);
sbi->nls = nls;
return 0;
out_put_hidden_dir:
iput(sbi->hidden_dir);
out_put_root:
iput(root);
out_put_alloc_file:
iput(sbi->alloc_file);
out_close_cat_tree:
hfs_btree_close(sbi->cat_tree);
out_close_ext_tree:
hfs_btree_close(sbi->ext_tree);
out_free_vhdr:
kfree(sbi->s_vhdr);
kfree(sbi->s_backup_vhdr);
out_unload_nls:
unload_nls(sbi->nls);
unload_nls(nls);
kfree(sbi);
out:
return err;
}
/*
* Allocate new inodes in the allocation group specified by agbp.
* Return 0 for success, else error code.
*/
STATIC int /* error code or 0 */
xfs_ialloc_ag_alloc(
xfs_trans_t *tp, /* transaction pointer */
xfs_buf_t *agbp, /* alloc group buffer */
int *alloc)
{
xfs_agi_t *agi; /* allocation group header */
xfs_alloc_arg_t args; /* allocation argument structure */
int blks_per_cluster; /* fs blocks per inode cluster */
xfs_btree_cur_t *cur; /* inode btree cursor */
xfs_daddr_t d; /* disk addr of buffer */
xfs_agnumber_t agno;
int error;
xfs_buf_t *fbuf; /* new free inodes' buffer */
xfs_dinode_t *free; /* new free inode structure */
int i; /* inode counter */
int j; /* block counter */
int nbufs; /* num bufs of new inodes */
xfs_agino_t newino; /* new first inode's number */
xfs_agino_t newlen; /* new number of inodes */
int ninodes; /* num inodes per buf */
xfs_agino_t thisino; /* current inode number, for loop */
int version; /* inode version number to use */
int isaligned = 0; /* inode allocation at stripe unit */
/* boundary */
unsigned int gen;
args.tp = tp;
args.mp = tp->t_mountp;
/*
* Locking will ensure that we don't have two callers in here
* at one time.
*/
newlen = XFS_IALLOC_INODES(args.mp);
if (args.mp->m_maxicount &&
args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
return XFS_ERROR(ENOSPC);
args.minlen = args.maxlen = XFS_IALLOC_BLOCKS(args.mp);
/*
* First try to allocate inodes contiguous with the last-allocated
* chunk of inodes. If the filesystem is striped, this will fill
* an entire stripe unit with inodes.
*/
agi = XFS_BUF_TO_AGI(agbp);
newino = be32_to_cpu(agi->agi_newino);
args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
XFS_IALLOC_BLOCKS(args.mp);
if (likely(newino != NULLAGINO &&
(args.agbno < be32_to_cpu(agi->agi_length)))) {
args.fsbno = XFS_AGB_TO_FSB(args.mp,
be32_to_cpu(agi->agi_seqno), args.agbno);
args.type = XFS_ALLOCTYPE_THIS_BNO;
args.mod = args.total = args.wasdel = args.isfl =
args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* We need to take into account alignment here to ensure that
* we don't modify the free list if we fail to have an exact
* block. If we don't have an exact match, and every oher
* attempt allocation attempt fails, we'll end up cancelling
* a dirty transaction and shutting down.
*
* For an exact allocation, alignment must be 1,
* however we need to take cluster alignment into account when
* fixing up the freelist. Use the minalignslop field to
* indicate that extra blocks might be required for alignment,
* but not to use them in the actual exact allocation.
*/
args.alignment = 1;
args.minalignslop = xfs_ialloc_cluster_alignment(&args) - 1;
/* Allow space for the inode btree to split. */
args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
if ((error = xfs_alloc_vextent(&args)))
return error;
} else
args.fsbno = NULLFSBLOCK;
if (unlikely(args.fsbno == NULLFSBLOCK)) {
/*
* Set the alignment for the allocation.
* If stripe alignment is turned on then align at stripe unit
* boundary.
* If the cluster size is smaller than a filesystem block
* then we're doing I/O for inodes in filesystem block size
* pieces, so don't need alignment anyway.
*/
isaligned = 0;
if (args.mp->m_sinoalign) {
ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN));
args.alignment = args.mp->m_dalign;
isaligned = 1;
} else
args.alignment = xfs_ialloc_cluster_alignment(&args);
/*
* Need to figure out where to allocate the inode blocks.
* Ideally they should be spaced out through the a.g.
* For now, just allocate blocks up front.
*/
args.agbno = be32_to_cpu(agi->agi_root);
args.fsbno = XFS_AGB_TO_FSB(args.mp,
be32_to_cpu(agi->agi_seqno), args.agbno);
/*
* Allocate a fixed-size extent of inodes.
*/
args.type = XFS_ALLOCTYPE_NEAR_BNO;
args.mod = args.total = args.wasdel = args.isfl =
args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* Allow space for the inode btree to split.
*/
args.minleft = XFS_IN_MAXLEVELS(args.mp) - 1;
if ((error = xfs_alloc_vextent(&args)))
return error;
}
/*
* If stripe alignment is turned on, then try again with cluster
* alignment.
*/
if (isaligned && args.fsbno == NULLFSBLOCK) {
args.type = XFS_ALLOCTYPE_NEAR_BNO;
args.agbno = be32_to_cpu(agi->agi_root);
args.fsbno = XFS_AGB_TO_FSB(args.mp,
be32_to_cpu(agi->agi_seqno), args.agbno);
args.alignment = xfs_ialloc_cluster_alignment(&args);
if ((error = xfs_alloc_vextent(&args)))
return error;
}
if (args.fsbno == NULLFSBLOCK) {
*alloc = 0;
return 0;
}
ASSERT(args.len == args.minlen);
/*
* Convert the results.
*/
newino = XFS_OFFBNO_TO_AGINO(args.mp, args.agbno, 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 (args.mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(args.mp)) {
blks_per_cluster = 1;
nbufs = (int)args.len;
ninodes = args.mp->m_sb.sb_inopblock;
} else {
blks_per_cluster = XFS_INODE_CLUSTER_SIZE(args.mp) /
args.mp->m_sb.sb_blocksize;
nbufs = (int)args.len / blks_per_cluster;
ninodes = blks_per_cluster * args.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.
*/
if (xfs_sb_version_hasnlink(&args.mp->m_sb))
version = XFS_DINODE_VERSION_2;
else
version = XFS_DINODE_VERSION_1;
/*
* Seed the new inode cluster with a random generation number. This
* prevents short-term reuse of generation numbers if a chunk is
* freed and then immediately reallocated. We use random numbers
* rather than a linear progression to prevent the next generation
* number from being easily guessable.
*/
gen = random32();
for (j = 0; j < nbufs; j++) {
/*
* Get the block.
*/
d = XFS_AGB_TO_DADDR(args.mp, be32_to_cpu(agi->agi_seqno),
args.agbno + (j * blks_per_cluster));
fbuf = xfs_trans_get_buf(tp, args.mp->m_ddev_targp, d,
args.mp->m_bsize * blks_per_cluster,
XFS_BUF_LOCK);
ASSERT(fbuf);
ASSERT(!XFS_BUF_GETERROR(fbuf));
/*
* Set initial values for the inodes in this buffer.
*/
xfs_biozero(fbuf, 0, ninodes << args.mp->m_sb.sb_inodelog);
for (i = 0; i < ninodes; i++) {
free = XFS_MAKE_IPTR(args.mp, fbuf, i);
free->di_core.di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
free->di_core.di_version = version;
free->di_core.di_gen = cpu_to_be32(gen);
free->di_next_unlinked = cpu_to_be32(NULLAGINO);
xfs_ialloc_log_di(tp, fbuf, i,
XFS_DI_CORE_BITS | XFS_DI_NEXT_UNLINKED);
}
xfs_trans_inode_alloc_buf(tp, fbuf);
}
be32_add_cpu(&agi->agi_count, newlen);
be32_add_cpu(&agi->agi_freecount, newlen);
agno = be32_to_cpu(agi->agi_seqno);
down_read(&args.mp->m_peraglock);
args.mp->m_perag[agno].pagi_freecount += newlen;
up_read(&args.mp->m_peraglock);
agi->agi_newino = cpu_to_be32(newino);
/*
* Insert records describing the new inode chunk into the btree.
*/
cur = xfs_btree_init_cursor(args.mp, tp, agbp, agno,
XFS_BTNUM_INO, (xfs_inode_t *)0, 0);
for (thisino = newino;
thisino < newino + newlen;
thisino += XFS_INODES_PER_CHUNK) {
if ((error = xfs_inobt_lookup_eq(cur, thisino,
XFS_INODES_PER_CHUNK, XFS_INOBT_ALL_FREE, &i))) {
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
return error;
}
ASSERT(i == 0);
if ((error = xfs_inobt_insert(cur, &i))) {
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
return error;
}
ASSERT(i == 1);
}
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
/*
* Log allocation group header fields
*/
xfs_ialloc_log_agi(tp, agbp,
XFS_AGI_COUNT | XFS_AGI_FREECOUNT | XFS_AGI_NEWINO);
/*
* Modify/log superblock values for inode count and inode free count.
*/
xfs_trans_mod_sb(tp, XFS_TRANS_SB_ICOUNT, (long)newlen);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, (long)newlen);
*alloc = 1;
return 0;
}
/*
* 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 new inodes in the allocation group specified by agbp.
* Return 0 for success, else error code.
*/
STATIC int /* error code or 0 */
xfs_ialloc_ag_alloc(
xfs_trans_t *tp, /* transaction pointer */
xfs_buf_t *agbp, /* alloc group buffer */
int *alloc)
{
xfs_agi_t *agi; /* allocation group header */
xfs_alloc_arg_t args; /* allocation argument structure */
xfs_btree_cur_t *cur; /* inode btree cursor */
xfs_agnumber_t agno;
int error;
int i;
xfs_agino_t newino; /* new first inode's number */
xfs_agino_t newlen; /* new number of inodes */
xfs_agino_t thisino; /* current inode number, for loop */
int isaligned = 0; /* inode allocation at stripe unit */
/* boundary */
struct xfs_perag *pag;
args.tp = tp;
args.mp = tp->t_mountp;
/*
* Locking will ensure that we don't have two callers in here
* at one time.
*/
newlen = XFS_IALLOC_INODES(args.mp);
if (args.mp->m_maxicount &&
args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
return XFS_ERROR(ENOSPC);
args.minlen = args.maxlen = XFS_IALLOC_BLOCKS(args.mp);
/*
* First try to allocate inodes contiguous with the last-allocated
* chunk of inodes. If the filesystem is striped, this will fill
* an entire stripe unit with inodes.
*/
agi = XFS_BUF_TO_AGI(agbp);
newino = be32_to_cpu(agi->agi_newino);
agno = be32_to_cpu(agi->agi_seqno);
args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
XFS_IALLOC_BLOCKS(args.mp);
if (likely(newino != NULLAGINO &&
(args.agbno < be32_to_cpu(agi->agi_length)))) {
args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
args.type = XFS_ALLOCTYPE_THIS_BNO;
args.mod = args.total = args.wasdel = args.isfl =
args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* We need to take into account alignment here to ensure that
* we don't modify the free list if we fail to have an exact
* block. If we don't have an exact match, and every oher
* attempt allocation attempt fails, we'll end up cancelling
* a dirty transaction and shutting down.
*
* For an exact allocation, alignment must be 1,
* however we need to take cluster alignment into account when
* fixing up the freelist. Use the minalignslop field to
* indicate that extra blocks might be required for alignment,
* but not to use them in the actual exact allocation.
*/
args.alignment = 1;
args.minalignslop = xfs_ialloc_cluster_alignment(&args) - 1;
/* Allow space for the inode btree to split. */
args.minleft = args.mp->m_in_maxlevels - 1;
if ((error = xfs_alloc_vextent(&args)))
return error;
} else
args.fsbno = NULLFSBLOCK;
if (unlikely(args.fsbno == NULLFSBLOCK)) {
/*
* Set the alignment for the allocation.
* If stripe alignment is turned on then align at stripe unit
* boundary.
* If the cluster size is smaller than a filesystem block
* then we're doing I/O for inodes in filesystem block size
* pieces, so don't need alignment anyway.
*/
isaligned = 0;
if (args.mp->m_sinoalign) {
ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN));
args.alignment = args.mp->m_dalign;
isaligned = 1;
} else
args.alignment = xfs_ialloc_cluster_alignment(&args);
/*
* Need to figure out where to allocate the inode blocks.
* Ideally they should be spaced out through the a.g.
* For now, just allocate blocks up front.
*/
args.agbno = be32_to_cpu(agi->agi_root);
args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
/*
* Allocate a fixed-size extent of inodes.
*/
args.type = XFS_ALLOCTYPE_NEAR_BNO;
args.mod = args.total = args.wasdel = args.isfl =
args.userdata = args.minalignslop = 0;
args.prod = 1;
/*
* Allow space for the inode btree to split.
*/
args.minleft = args.mp->m_in_maxlevels - 1;
if ((error = xfs_alloc_vextent(&args)))
return error;
}
/*
* If stripe alignment is turned on, then try again with cluster
* alignment.
*/
if (isaligned && args.fsbno == NULLFSBLOCK) {
args.type = XFS_ALLOCTYPE_NEAR_BNO;
args.agbno = be32_to_cpu(agi->agi_root);
args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
args.alignment = xfs_ialloc_cluster_alignment(&args);
if ((error = xfs_alloc_vextent(&args)))
return error;
}
if (args.fsbno == NULLFSBLOCK) {
*alloc = 0;
return 0;
}
ASSERT(args.len == args.minlen);
/*
* Stamp and write the inode buffers.
*
* Seed the new inode cluster with a random generation number. This
* prevents short-term reuse of generation numbers if a chunk is
* freed and then immediately reallocated. We use random numbers
* rather than a linear progression to prevent the next generation
* number from being easily guessable.
*/
error = xfs_ialloc_inode_init(args.mp, tp, agno, args.agbno,
args.len, prandom_u32());
if (error)
return error;
/*
* Convert the results.
*/
newino = XFS_OFFBNO_TO_AGINO(args.mp, args.agbno, 0);
be32_add_cpu(&agi->agi_count, newlen);
be32_add_cpu(&agi->agi_freecount, newlen);
pag = xfs_perag_get(args.mp, agno);
pag->pagi_freecount += newlen;
xfs_perag_put(pag);
agi->agi_newino = cpu_to_be32(newino);
/*
* Insert records describing the new inode chunk into the btree.
*/
cur = xfs_inobt_init_cursor(args.mp, tp, agbp, agno);
for (thisino = newino;
thisino < newino + newlen;
thisino += XFS_INODES_PER_CHUNK) {
cur->bc_rec.i.ir_startino = thisino;
cur->bc_rec.i.ir_freecount = XFS_INODES_PER_CHUNK;
cur->bc_rec.i.ir_free = XFS_INOBT_ALL_FREE;
error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, &i);
if (error) {
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
return error;
}
ASSERT(i == 0);
error = xfs_btree_insert(cur, &i);
if (error) {
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
return error;
}
ASSERT(i == 1);
}
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
/*
* Log allocation group header fields
*/
xfs_ialloc_log_agi(tp, agbp,
XFS_AGI_COUNT | XFS_AGI_FREECOUNT | XFS_AGI_NEWINO);
/*
* Modify/log superblock values for inode count and inode free count.
*/
xfs_trans_mod_sb(tp, XFS_TRANS_SB_ICOUNT, (long)newlen);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, (long)newlen);
*alloc = 1;
return 0;
}
/*
* 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;
}
/*
* xfs_trans_apply_sb_deltas() is called from the commit code
* to bring the superblock buffer into the current transaction
* and modify it as requested by earlier calls to xfs_trans_mod_sb().
*
* For now we just look at each field allowed to change and change
* it if necessary.
*/
STATIC void
xfs_trans_apply_sb_deltas(
xfs_trans_t *tp)
{
xfs_dsb_t *sbp;
xfs_buf_t *bp;
int whole = 0;
bp = xfs_trans_getsb(tp, tp->t_mountp, 0);
sbp = XFS_BUF_TO_SBP(bp);
/*
* Check that superblock mods match the mods made to AGF counters.
*/
ASSERT((tp->t_fdblocks_delta + tp->t_res_fdblocks_delta) ==
(tp->t_ag_freeblks_delta + tp->t_ag_flist_delta +
tp->t_ag_btree_delta));
/*
* Only update the superblock counters if we are logging them
*/
if (!xfs_sb_version_haslazysbcount(&(tp->t_mountp->m_sb))) {
if (tp->t_icount_delta)
be64_add_cpu(&sbp->sb_icount, tp->t_icount_delta);
if (tp->t_ifree_delta)
be64_add_cpu(&sbp->sb_ifree, tp->t_ifree_delta);
if (tp->t_fdblocks_delta)
be64_add_cpu(&sbp->sb_fdblocks, tp->t_fdblocks_delta);
if (tp->t_res_fdblocks_delta)
be64_add_cpu(&sbp->sb_fdblocks, tp->t_res_fdblocks_delta);
}
if (tp->t_frextents_delta)
be64_add_cpu(&sbp->sb_frextents, tp->t_frextents_delta);
if (tp->t_res_frextents_delta)
be64_add_cpu(&sbp->sb_frextents, tp->t_res_frextents_delta);
if (tp->t_dblocks_delta) {
be64_add_cpu(&sbp->sb_dblocks, tp->t_dblocks_delta);
whole = 1;
}
if (tp->t_agcount_delta) {
be32_add_cpu(&sbp->sb_agcount, tp->t_agcount_delta);
whole = 1;
}
if (tp->t_imaxpct_delta) {
sbp->sb_imax_pct += tp->t_imaxpct_delta;
whole = 1;
}
if (tp->t_rextsize_delta) {
be32_add_cpu(&sbp->sb_rextsize, tp->t_rextsize_delta);
whole = 1;
}
if (tp->t_rbmblocks_delta) {
be32_add_cpu(&sbp->sb_rbmblocks, tp->t_rbmblocks_delta);
whole = 1;
}
if (tp->t_rblocks_delta) {
be64_add_cpu(&sbp->sb_rblocks, tp->t_rblocks_delta);
whole = 1;
}
if (tp->t_rextents_delta) {
be64_add_cpu(&sbp->sb_rextents, tp->t_rextents_delta);
whole = 1;
}
if (tp->t_rextslog_delta) {
sbp->sb_rextslog += tp->t_rextslog_delta;
whole = 1;
}
if (whole)
/*
* Log the whole thing, the fields are noncontiguous.
*/
xfs_trans_log_buf(tp, bp, 0, sizeof(xfs_dsb_t) - 1);
else
/*
* Since all the modifiable fields are contiguous, we
* can get away with this.
*/
xfs_trans_log_buf(tp, bp, offsetof(xfs_dsb_t, sb_icount),
offsetof(xfs_dsb_t, sb_frextents) +
sizeof(sbp->sb_frextents) - 1);
tp->t_mountp->m_super->s_dirt = 1;
}
/*
* Add an entry to a block directory.
*/
int /* error */
xfs_dir2_block_addname(
xfs_da_args_t *args) /* directory op arguments */
{
xfs_dir2_data_free_t *bf; /* bestfree table in block */
xfs_dir2_block_t *block; /* directory block structure */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
xfs_dabuf_t *bp; /* buffer for block */
xfs_dir2_block_tail_t *btp; /* block tail */
int compact; /* need to compact leaf ents */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* directory inode */
xfs_dir2_data_unused_t *dup; /* block unused entry */
int error; /* error return value */
xfs_dir2_data_unused_t *enddup=NULL; /* unused at end of data */
xfs_dahash_t hash; /* hash value of found entry */
int high; /* high index for binary srch */
int highstale; /* high stale index */
int lfloghigh=0; /* last final leaf to log */
int lfloglow=0; /* first final leaf to log */
int len; /* length of the new entry */
int low; /* low index for binary srch */
int lowstale; /* low stale index */
int mid=0; /* midpoint for binary srch */
xfs_mount_t *mp; /* filesystem mount point */
int needlog; /* need to log header */
int needscan; /* need to rescan freespace */
__be16 *tagp; /* pointer to tag value */
xfs_trans_t *tp; /* transaction structure */
trace_xfs_dir2_block_addname(args);
dp = args->dp;
tp = args->trans;
mp = dp->i_mount;
/*
* Read the (one and only) directory block into dabuf bp.
*/
if ((error =
xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, &bp, XFS_DATA_FORK))) {
return error;
}
ASSERT(bp != NULL);
block = bp->data;
/*
* Check the magic number, corrupted if wrong.
*/
if (unlikely(be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)) {
XFS_CORRUPTION_ERROR("xfs_dir2_block_addname",
XFS_ERRLEVEL_LOW, mp, block);
xfs_da_brelse(tp, bp);
return XFS_ERROR(EFSCORRUPTED);
}
len = xfs_dir2_data_entsize(args->namelen);
/*
* Set up pointers to parts of the block.
*/
bf = block->hdr.bestfree;
btp = xfs_dir2_block_tail_p(mp, block);
blp = xfs_dir2_block_leaf_p(btp);
/*
* No stale entries? Need space for entry and new leaf.
*/
if (!btp->stale) {
/*
* Tag just before the first leaf entry.
*/
tagp = (__be16 *)blp - 1;
/*
* Data object just before the first leaf entry.
*/
enddup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
/*
* If it's not free then can't do this add without cleaning up:
* the space before the first leaf entry needs to be free so it
* can be expanded to hold the pointer to the new entry.
*/
if (be16_to_cpu(enddup->freetag) != XFS_DIR2_DATA_FREE_TAG)
dup = enddup = NULL;
/*
* Check out the biggest freespace and see if it's the same one.
*/
else {
dup = (xfs_dir2_data_unused_t *)
((char *)block + be16_to_cpu(bf[0].offset));
if (dup == enddup) {
/*
* It is the biggest freespace, is it too small
* to hold the new leaf too?
*/
if (be16_to_cpu(dup->length) < len + (uint)sizeof(*blp)) {
/*
* Yes, we use the second-largest
* entry instead if it works.
*/
if (be16_to_cpu(bf[1].length) >= len)
dup = (xfs_dir2_data_unused_t *)
((char *)block +
be16_to_cpu(bf[1].offset));
else
dup = NULL;
}
} else {
/*
* Not the same free entry,
* just check its length.
*/
if (be16_to_cpu(dup->length) < len) {
dup = NULL;
}
}
}
compact = 0;
}
/*
* If there are stale entries we'll use one for the leaf.
* Is the biggest entry enough to avoid compaction?
*/
else if (be16_to_cpu(bf[0].length) >= len) {
dup = (xfs_dir2_data_unused_t *)
((char *)block + be16_to_cpu(bf[0].offset));
compact = 0;
}
/*
* Will need to compact to make this work.
*/
else {
/*
* Tag just before the first leaf entry.
*/
tagp = (__be16 *)blp - 1;
/*
* Data object just before the first leaf entry.
*/
dup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
/*
* If it's not free then the data will go where the
* leaf data starts now, if it works at all.
*/
if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
if (be16_to_cpu(dup->length) + (be32_to_cpu(btp->stale) - 1) *
(uint)sizeof(*blp) < len)
dup = NULL;
} else if ((be32_to_cpu(btp->stale) - 1) * (uint)sizeof(*blp) < len)
dup = NULL;
else
dup = (xfs_dir2_data_unused_t *)blp;
compact = 1;
}
/*
* If this isn't a real add, we're done with the buffer.
*/
if (args->op_flags & XFS_DA_OP_JUSTCHECK)
xfs_da_brelse(tp, bp);
/*
* If we don't have space for the new entry & leaf ...
*/
if (!dup) {
/*
* Not trying to actually do anything, or don't have
* a space reservation: return no-space.
*/
if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
return XFS_ERROR(ENOSPC);
/*
* Convert to the next larger format.
* Then add the new entry in that format.
*/
error = xfs_dir2_block_to_leaf(args, bp);
xfs_da_buf_done(bp);
if (error)
return error;
return xfs_dir2_leaf_addname(args);
}
/*
* Just checking, and it would work, so say so.
*/
if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
needlog = needscan = 0;
/*
* If need to compact the leaf entries, do it now.
* Leave the highest-numbered stale entry stale.
* XXX should be the one closest to mid but mid is not yet computed.
*/
if (compact) {
int fromidx; /* source leaf index */
int toidx; /* target leaf index */
for (fromidx = toidx = be32_to_cpu(btp->count) - 1,
highstale = lfloghigh = -1;
fromidx >= 0;
fromidx--) {
if (be32_to_cpu(blp[fromidx].address) == XFS_DIR2_NULL_DATAPTR) {
if (highstale == -1)
highstale = toidx;
else {
if (lfloghigh == -1)
lfloghigh = toidx;
continue;
}
}
if (fromidx < toidx)
blp[toidx] = blp[fromidx];
toidx--;
}
lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
lfloghigh -= be32_to_cpu(btp->stale) - 1;
be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
xfs_dir2_data_make_free(tp, bp,
(xfs_dir2_data_aoff_t)((char *)blp - (char *)block),
(xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
&needlog, &needscan);
blp += be32_to_cpu(btp->stale) - 1;
btp->stale = cpu_to_be32(1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
*/
if (needscan) {
xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
needscan = 0;
}
}
/*
* Set leaf logging boundaries to impossible state.
* For the no-stale case they're set explicitly.
*/
else if (btp->stale) {
lfloglow = be32_to_cpu(btp->count);
lfloghigh = -1;
}
/*
* Find the slot that's first lower than our hash value, -1 if none.
*/
for (low = 0, high = be32_to_cpu(btp->count) - 1; low <= high; ) {
mid = (low + high) >> 1;
if ((hash = be32_to_cpu(blp[mid].hashval)) == args->hashval)
break;
if (hash < args->hashval)
low = mid + 1;
else
high = mid - 1;
}
while (mid >= 0 && be32_to_cpu(blp[mid].hashval) >= args->hashval) {
mid--;
}
/*
* No stale entries, will use enddup space to hold new leaf.
*/
if (!btp->stale) {
/*
* Mark the space needed for the new leaf entry, now in use.
*/
xfs_dir2_data_use_free(tp, bp, enddup,
(xfs_dir2_data_aoff_t)
((char *)enddup - (char *)block + be16_to_cpu(enddup->length) -
sizeof(*blp)),
(xfs_dir2_data_aoff_t)sizeof(*blp),
&needlog, &needscan);
/*
* Update the tail (entry count).
*/
be32_add_cpu(&btp->count, 1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
*/
if (needscan) {
xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block,
&needlog);
needscan = 0;
}
/*
* Adjust pointer to the first leaf entry, we're about to move
* the table up one to open up space for the new leaf entry.
* Then adjust our index to match.
*/
blp--;
mid++;
if (mid)
memmove(blp, &blp[1], mid * sizeof(*blp));
lfloglow = 0;
lfloghigh = mid;
}
/*
* Use a stale leaf for our new entry.
*/
else {
for (lowstale = mid;
lowstale >= 0 &&
be32_to_cpu(blp[lowstale].address) != XFS_DIR2_NULL_DATAPTR;
lowstale--)
continue;
for (highstale = mid + 1;
highstale < be32_to_cpu(btp->count) &&
be32_to_cpu(blp[highstale].address) != XFS_DIR2_NULL_DATAPTR &&
(lowstale < 0 || mid - lowstale > highstale - mid);
highstale++)
continue;
/*
* Move entries toward the low-numbered stale entry.
*/
if (lowstale >= 0 &&
(highstale == be32_to_cpu(btp->count) ||
mid - lowstale <= highstale - mid)) {
if (mid - lowstale)
memmove(&blp[lowstale], &blp[lowstale + 1],
(mid - lowstale) * sizeof(*blp));
lfloglow = MIN(lowstale, lfloglow);
lfloghigh = MAX(mid, lfloghigh);
}
/*
* Move entries toward the high-numbered stale entry.
*/
else {
ASSERT(highstale < be32_to_cpu(btp->count));
mid++;
if (highstale - mid)
memmove(&blp[mid + 1], &blp[mid],
(highstale - mid) * sizeof(*blp));
lfloglow = MIN(mid, lfloglow);
lfloghigh = MAX(highstale, lfloghigh);
}
be32_add_cpu(&btp->stale, -1);
}
/*
* Point to the new data entry.
*/
dep = (xfs_dir2_data_entry_t *)dup;
/*
* Fill in the leaf entry.
*/
blp[mid].hashval = cpu_to_be32(args->hashval);
blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
(char *)dep - (char *)block));
xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh);
/*
* Mark space for the data entry used.
*/
xfs_dir2_data_use_free(tp, bp, dup,
(xfs_dir2_data_aoff_t)((char *)dup - (char *)block),
(xfs_dir2_data_aoff_t)len, &needlog, &needscan);
/*
* Create the new data entry.
*/
dep->inumber = cpu_to_be64(args->inumber);
dep->namelen = args->namelen;
memcpy(dep->name, args->name, args->namelen);
tagp = xfs_dir2_data_entry_tag_p(dep);
*tagp = cpu_to_be16((char *)dep - (char *)block);
/*
* Clean up the bestfree array and log the header, tail, and entry.
*/
if (needscan)
xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog);
if (needlog)
xfs_dir2_data_log_header(tp, bp);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_log_entry(tp, bp, dep);
xfs_dir2_data_check(dp, bp);
xfs_da_buf_done(bp);
return 0;
}
/*
* Add an entry to a block directory.
*/
int /* error */
xfs_dir2_block_addname(
xfs_da_args_t *args) /* directory op arguments */
{
xfs_dir2_data_hdr_t *hdr; /* block header */
xfs_dir2_leaf_entry_t *blp; /* block leaf entries */
struct xfs_buf *bp; /* buffer for block */
xfs_dir2_block_tail_t *btp; /* block tail */
int compact; /* need to compact leaf ents */
xfs_dir2_data_entry_t *dep; /* block data entry */
xfs_inode_t *dp; /* directory inode */
xfs_dir2_data_unused_t *dup; /* block unused entry */
int error; /* error return value */
xfs_dir2_data_unused_t *enddup=NULL; /* unused at end of data */
xfs_dahash_t hash; /* hash value of found entry */
int high; /* high index for binary srch */
int highstale; /* high stale index */
int lfloghigh=0; /* last final leaf to log */
int lfloglow=0; /* first final leaf to log */
int len; /* length of the new entry */
int low; /* low index for binary srch */
int lowstale; /* low stale index */
int mid=0; /* midpoint for binary srch */
int needlog; /* need to log header */
int needscan; /* need to rescan freespace */
__be16 *tagp; /* pointer to tag value */
xfs_trans_t *tp; /* transaction structure */
trace_xfs_dir2_block_addname(args);
dp = args->dp;
tp = args->trans;
/* Read the (one and only) directory block into bp. */
error = xfs_dir3_block_read(tp, dp, &bp);
if (error)
return error;
len = dp->d_ops->data_entsize(args->namelen);
/*
* Set up pointers to parts of the block.
*/
hdr = bp->b_addr;
btp = xfs_dir2_block_tail_p(args->geo, hdr);
blp = xfs_dir2_block_leaf_p(btp);
/*
* Find out if we can reuse stale entries or whether we need extra
* space for entry and new leaf.
*/
xfs_dir2_block_need_space(dp, hdr, btp, blp, &tagp, &dup,
&enddup, &compact, len);
/*
* Done everything we need for a space check now.
*/
if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
xfs_trans_brelse(tp, bp);
if (!dup)
return -ENOSPC;
return 0;
}
/*
* If we don't have space for the new entry & leaf ...
*/
if (!dup) {
/* Don't have a space reservation: return no-space. */
if (args->total == 0)
return -ENOSPC;
/*
* Convert to the next larger format.
* Then add the new entry in that format.
*/
error = xfs_dir2_block_to_leaf(args, bp);
if (error)
return error;
return xfs_dir2_leaf_addname(args);
}
needlog = needscan = 0;
/*
* If need to compact the leaf entries, do it now.
*/
if (compact) {
xfs_dir2_block_compact(args, bp, hdr, btp, blp, &needlog,
&lfloghigh, &lfloglow);
/* recalculate blp post-compaction */
blp = xfs_dir2_block_leaf_p(btp);
} else if (btp->stale) {
/*
* Set leaf logging boundaries to impossible state.
* For the no-stale case they're set explicitly.
*/
lfloglow = be32_to_cpu(btp->count);
lfloghigh = -1;
}
/*
* Find the slot that's first lower than our hash value, -1 if none.
*/
for (low = 0, high = be32_to_cpu(btp->count) - 1; low <= high; ) {
mid = (low + high) >> 1;
if ((hash = be32_to_cpu(blp[mid].hashval)) == args->hashval)
break;
if (hash < args->hashval)
low = mid + 1;
else
high = mid - 1;
}
while (mid >= 0 && be32_to_cpu(blp[mid].hashval) >= args->hashval) {
mid--;
}
/*
* No stale entries, will use enddup space to hold new leaf.
*/
if (!btp->stale) {
/*
* Mark the space needed for the new leaf entry, now in use.
*/
xfs_dir2_data_use_free(args, bp, enddup,
(xfs_dir2_data_aoff_t)
((char *)enddup - (char *)hdr + be16_to_cpu(enddup->length) -
sizeof(*blp)),
(xfs_dir2_data_aoff_t)sizeof(*blp),
&needlog, &needscan);
/*
* Update the tail (entry count).
*/
be32_add_cpu(&btp->count, 1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
*/
if (needscan) {
xfs_dir2_data_freescan(dp, hdr, &needlog);
needscan = 0;
}
/*
* Adjust pointer to the first leaf entry, we're about to move
* the table up one to open up space for the new leaf entry.
* Then adjust our index to match.
*/
blp--;
mid++;
if (mid)
memmove(blp, &blp[1], mid * sizeof(*blp));
lfloglow = 0;
lfloghigh = mid;
}
/*
* Use a stale leaf for our new entry.
*/
else {
for (lowstale = mid;
lowstale >= 0 &&
blp[lowstale].address !=
cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
lowstale--)
continue;
for (highstale = mid + 1;
highstale < be32_to_cpu(btp->count) &&
blp[highstale].address !=
cpu_to_be32(XFS_DIR2_NULL_DATAPTR) &&
(lowstale < 0 || mid - lowstale > highstale - mid);
highstale++)
continue;
/*
* Move entries toward the low-numbered stale entry.
*/
if (lowstale >= 0 &&
(highstale == be32_to_cpu(btp->count) ||
mid - lowstale <= highstale - mid)) {
if (mid - lowstale)
memmove(&blp[lowstale], &blp[lowstale + 1],
(mid - lowstale) * sizeof(*blp));
lfloglow = MIN(lowstale, lfloglow);
lfloghigh = MAX(mid, lfloghigh);
}
/*
* Move entries toward the high-numbered stale entry.
*/
else {
ASSERT(highstale < be32_to_cpu(btp->count));
mid++;
if (highstale - mid)
memmove(&blp[mid + 1], &blp[mid],
(highstale - mid) * sizeof(*blp));
lfloglow = MIN(mid, lfloglow);
lfloghigh = MAX(highstale, lfloghigh);
}
be32_add_cpu(&btp->stale, -1);
}
/*
* Point to the new data entry.
*/
dep = (xfs_dir2_data_entry_t *)dup;
/*
* Fill in the leaf entry.
*/
blp[mid].hashval = cpu_to_be32(args->hashval);
blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
(char *)dep - (char *)hdr));
xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh);
/*
* Mark space for the data entry used.
*/
xfs_dir2_data_use_free(args, bp, dup,
(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
(xfs_dir2_data_aoff_t)len, &needlog, &needscan);
/*
* Create the new data entry.
*/
dep->inumber = cpu_to_be64(args->inumber);
dep->namelen = args->namelen;
memcpy(dep->name, args->name, args->namelen);
dp->d_ops->data_put_ftype(dep, args->filetype);
tagp = dp->d_ops->data_entry_tag_p(dep);
*tagp = cpu_to_be16((char *)dep - (char *)hdr);
/*
* Clean up the bestfree array and log the header, tail, and entry.
*/
if (needscan)
xfs_dir2_data_freescan(dp, hdr, &needlog);
if (needlog)
xfs_dir2_data_log_header(args, bp);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_log_entry(args, bp, dep);
xfs_dir3_data_check(dp, bp);
return 0;
}
/*
* hfs_mdb_get()
*
* Build the in-core MDB for a filesystem, including
* the B-trees and the volume bitmap.
*/
int hfs_mdb_get(struct super_block *sb)
{
struct buffer_head *bh;
struct hfs_mdb *mdb, *mdb2;
unsigned int block;
char *ptr;
int off2, len, size, sect;
sector_t part_start, part_size;
loff_t off;
__be16 attrib;
/* set the device driver to 512-byte blocks */
size = sb_min_blocksize(sb, HFS_SECTOR_SIZE);
if (!size)
return -EINVAL;
if (hfs_get_last_session(sb, &part_start, &part_size))
return -EINVAL;
while (1) {
/* See if this is an HFS filesystem */
bh = sb_bread512(sb, part_start + HFS_MDB_BLK, mdb);
if (!bh)
goto out;
if (mdb->drSigWord == cpu_to_be16(HFS_SUPER_MAGIC))
break;
brelse(bh);
/* check for a partition block
* (should do this only for cdrom/loop though)
*/
if (hfs_part_find(sb, &part_start, &part_size))
goto out;
}
HFS_SB(sb)->alloc_blksz = size = be32_to_cpu(mdb->drAlBlkSiz);
if (!size || (size & (HFS_SECTOR_SIZE - 1))) {
pr_err("bad allocation block size %d\n", size);
goto out_bh;
}
size = min(HFS_SB(sb)->alloc_blksz, (u32)PAGE_SIZE);
/* size must be a multiple of 512 */
while (size & (size - 1))
size -= HFS_SECTOR_SIZE;
sect = be16_to_cpu(mdb->drAlBlSt) + part_start;
/* align block size to first sector */
while (sect & ((size - 1) >> HFS_SECTOR_SIZE_BITS))
size >>= 1;
/* align block size to weird alloc size */
while (HFS_SB(sb)->alloc_blksz & (size - 1))
size >>= 1;
brelse(bh);
if (!sb_set_blocksize(sb, size)) {
pr_err("unable to set blocksize to %u\n", size);
goto out;
}
bh = sb_bread512(sb, part_start + HFS_MDB_BLK, mdb);
if (!bh)
goto out;
if (mdb->drSigWord != cpu_to_be16(HFS_SUPER_MAGIC))
goto out_bh;
HFS_SB(sb)->mdb_bh = bh;
HFS_SB(sb)->mdb = mdb;
/* These parameters are read from the MDB, and never written */
HFS_SB(sb)->part_start = part_start;
HFS_SB(sb)->fs_ablocks = be16_to_cpu(mdb->drNmAlBlks);
HFS_SB(sb)->fs_div = HFS_SB(sb)->alloc_blksz >> sb->s_blocksize_bits;
HFS_SB(sb)->clumpablks = be32_to_cpu(mdb->drClpSiz) /
HFS_SB(sb)->alloc_blksz;
if (!HFS_SB(sb)->clumpablks)
HFS_SB(sb)->clumpablks = 1;
HFS_SB(sb)->fs_start = (be16_to_cpu(mdb->drAlBlSt) + part_start) >>
(sb->s_blocksize_bits - HFS_SECTOR_SIZE_BITS);
/* These parameters are read from and written to the MDB */
HFS_SB(sb)->free_ablocks = be16_to_cpu(mdb->drFreeBks);
HFS_SB(sb)->next_id = be32_to_cpu(mdb->drNxtCNID);
HFS_SB(sb)->root_files = be16_to_cpu(mdb->drNmFls);
HFS_SB(sb)->root_dirs = be16_to_cpu(mdb->drNmRtDirs);
HFS_SB(sb)->file_count = be32_to_cpu(mdb->drFilCnt);
HFS_SB(sb)->folder_count = be32_to_cpu(mdb->drDirCnt);
/* TRY to get the alternate (backup) MDB. */
sect = part_start + part_size - 2;
bh = sb_bread512(sb, sect, mdb2);
if (bh) {
if (mdb2->drSigWord == cpu_to_be16(HFS_SUPER_MAGIC)) {
HFS_SB(sb)->alt_mdb_bh = bh;
HFS_SB(sb)->alt_mdb = mdb2;
} else
brelse(bh);
}
if (!HFS_SB(sb)->alt_mdb) {
pr_warn("unable to locate alternate MDB\n");
pr_warn("continuing without an alternate MDB\n");
}
HFS_SB(sb)->bitmap = (__be32 *)__get_free_pages(GFP_KERNEL, PAGE_SIZE < 8192 ? 1 : 0);
if (!HFS_SB(sb)->bitmap)
goto out;
/* read in the bitmap */
block = be16_to_cpu(mdb->drVBMSt) + part_start;
off = (loff_t)block << HFS_SECTOR_SIZE_BITS;
size = (HFS_SB(sb)->fs_ablocks + 8) / 8;
ptr = (u8 *)HFS_SB(sb)->bitmap;
while (size) {
bh = sb_bread(sb, off >> sb->s_blocksize_bits);
if (!bh) {
pr_err("unable to read volume bitmap\n");
goto out;
}
off2 = off & (sb->s_blocksize - 1);
len = min((int)sb->s_blocksize - off2, size);
memcpy(ptr, bh->b_data + off2, len);
brelse(bh);
ptr += len;
off += len;
size -= len;
}
HFS_SB(sb)->ext_tree = hfs_btree_open(sb, HFS_EXT_CNID, hfs_ext_keycmp);
if (!HFS_SB(sb)->ext_tree) {
pr_err("unable to open extent tree\n");
goto out;
}
HFS_SB(sb)->cat_tree = hfs_btree_open(sb, HFS_CAT_CNID, hfs_cat_keycmp);
if (!HFS_SB(sb)->cat_tree) {
pr_err("unable to open catalog tree\n");
goto out;
}
attrib = mdb->drAtrb;
if (!(attrib & cpu_to_be16(HFS_SB_ATTRIB_UNMNT))) {
pr_warn("filesystem was not cleanly unmounted, running fsck.hfs is recommended. mounting read-only.\n");
sb->s_flags |= MS_RDONLY;
}
if ((attrib & cpu_to_be16(HFS_SB_ATTRIB_SLOCK))) {
pr_warn("filesystem is marked locked, mounting read-only.\n");
sb->s_flags |= MS_RDONLY;
}
if (!(sb->s_flags & MS_RDONLY)) {
/* Mark the volume uncleanly unmounted in case we crash */
attrib &= cpu_to_be16(~HFS_SB_ATTRIB_UNMNT);
attrib |= cpu_to_be16(HFS_SB_ATTRIB_INCNSTNT);
mdb->drAtrb = attrib;
be32_add_cpu(&mdb->drWrCnt, 1);
mdb->drLsMod = hfs_mtime();
mark_buffer_dirty(HFS_SB(sb)->mdb_bh);
sync_dirty_buffer(HFS_SB(sb)->mdb_bh);
}
return 0;
out_bh:
brelse(bh);
out:
hfs_mdb_put(sb);
return -EIO;
}
/*
* 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;
}