int
xfs_discard_extents(
struct xfs_mount *mp,
struct list_head *list)
{
struct xfs_busy_extent *busyp;
int error = 0;
list_for_each_entry(busyp, list, list) {
trace_xfs_discard_extent(mp, busyp->agno, busyp->bno,
busyp->length);
error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno),
XFS_FSB_TO_BB(mp, busyp->length),
GFP_NOFS, 0);
if (error && error != EOPNOTSUPP) {
xfs_info(mp,
"discard failed for extent [0x%llu,%u], error %d",
(unsigned long long)busyp->bno,
busyp->length,
error);
return error;
}
}
static void
scanfunc_ino(
struct xfs_btree_block *block,
int level,
xfs_agf_t *agf)
{
xfs_agino_t agino;
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
int i;
int j;
int off;
xfs_inobt_ptr_t *pp;
xfs_inobt_rec_t *rp;
if (level == 0) {
rp = XFS_INOBT_REC_ADDR(mp, block, 1);
for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++) {
agino = be32_to_cpu(rp[i].ir_startino);
off = XFS_INO_TO_OFFSET(mp, agino);
push_cur();
set_cur(&typtab[TYP_INODE],
XFS_AGB_TO_DADDR(mp, seqno,
XFS_AGINO_TO_AGBNO(mp, agino)),
XFS_FSB_TO_BB(mp, XFS_IALLOC_BLOCKS(mp)),
DB_RING_IGN, NULL);
if (iocur_top->data == NULL) {
dbprintf(_("can't read inode block %u/%u\n"),
seqno, XFS_AGINO_TO_AGBNO(mp, agino));
continue;
}
for (j = 0; j < XFS_INODES_PER_CHUNK; j++) {
if (XFS_INOBT_IS_FREE_DISK(&rp[i], j))
continue;
process_inode(agf, agino + j, (xfs_dinode_t *)
((char *)iocur_top->data +
((off + j) << mp->m_sb.sb_inodelog)));
}
pop_cur();
}
return;
}
pp = XFS_INOBT_PTR_ADDR(mp, block, 1, mp->m_inobt_mxr[1]);
for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++)
scan_sbtree(agf, be32_to_cpu(pp[i]), level, scanfunc_ino,
TYP_INOBT);
}
static void
scan_sbtree(
xfs_agf_t *agf,
xfs_agblock_t root,
int nlevels,
scan_sbtree_f_t func,
typnm_t btype)
{
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
push_cur();
set_cur(&typtab[btype], XFS_AGB_TO_DADDR(mp, seqno, root),
blkbb, DB_RING_IGN, NULL);
if (iocur_top->data == NULL) {
dbprintf(_("can't read btree block %u/%u\n"), seqno, root);
return;
}
(*func)(iocur_top->data, nlevels - 1, agf);
pop_cur();
}
/* Transform a bnobt irec into a fsmap */
STATIC int
xfs_getfsmap_datadev_bnobt_helper(
struct xfs_btree_cur *cur,
struct xfs_alloc_rec_incore *rec,
void *priv)
{
struct xfs_mount *mp = cur->bc_mp;
struct xfs_getfsmap_info *info = priv;
struct xfs_rmap_irec irec;
xfs_daddr_t rec_daddr;
rec_daddr = XFS_AGB_TO_DADDR(mp, cur->bc_private.a.agno,
rec->ar_startblock);
irec.rm_startblock = rec->ar_startblock;
irec.rm_blockcount = rec->ar_blockcount;
irec.rm_owner = XFS_RMAP_OWN_NULL; /* "free" */
irec.rm_offset = 0;
irec.rm_flags = 0;
return xfs_getfsmap_helper(cur->bc_tp, info, &irec, rec_daddr);
}
/*
* 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;
/*
* 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.
*/
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_LOCK);
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.
*/
xfs_buf_zero(fbuf, 0, ninodes << mp->m_sb.sb_inodelog);
for (i = 0; i < ninodes; i++) {
int ioffset = i << mp->m_sb.sb_inodelog;
uint isize = sizeof(struct xfs_dinode);
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);
xfs_trans_log_buf(tp, fbuf, ioffset, ioffset + isize - 1);
}
xfs_trans_inode_alloc_buf(tp, fbuf);
}
return 0;
}
extern void readbitmap(char* device, image_head image_hdr, unsigned long* bitmap, int pui)
{
xfs_agnumber_t agno = 0;
xfs_agblock_t first_agbno;
xfs_agnumber_t num_ags;
ag_header_t ag_hdr;
xfs_daddr_t read_ag_off;
int read_ag_length;
void *read_ag_buf = NULL;
xfs_off_t read_ag_position; /* xfs_types.h: typedef __s64 */
uint64_t sk, res, s_pos = 0;
void *btree_buf_data = NULL;
int btree_buf_length;
xfs_off_t btree_buf_position;
xfs_agblock_t bno;
uint current_level;
uint btree_levels;
xfs_daddr_t begin, next_begin, ag_begin, new_begin, ag_end;
/* xfs_types.h: typedef __s64*/
xfs_off_t pos;
xfs_alloc_ptr_t *ptr;
xfs_alloc_rec_t *rec_ptr;
int length;
int i;
uint64_t size, sizeb;
xfs_off_t w_position;
int w_length;
int wblocks;
int w_size = 1 * 1024 * 1024;
uint64_t numblocks = 0;
xfs_off_t logstart, logend;
xfs_off_t logstart_pos, logend_pos;
int log_length;
struct xfs_btree_block *block;
uint64_t current_block, block_count, prog_cur_block = 0;
int start = 0;
int bit_size = 1;
progress_bar prog;
uint64_t bused = 0;
uint64_t bfree = 0;
/// init progress
progress_init(&prog, start, image_hdr.totalblock, image_hdr.totalblock, BITMAP, bit_size);
fs_open(device);
first_agbno = (((XFS_AGFL_DADDR(mp) + 1) * source_sectorsize) + first_residue) / source_blocksize;
num_ags = mp->m_sb.sb_agcount;
log_mesg(1, 0, 0, fs_opt.debug, "ags = %i\n", num_ags);
for (agno = 0; agno < num_ags ; agno++) {
/* read in first blocks of the ag */
/* initial settings */
log_mesg(2, 0, 0, fs_opt.debug, "read ag %i header\n", agno);
read_ag_off = XFS_AG_DADDR(mp, agno, XFS_SB_DADDR);
read_ag_length = first_agbno * source_blocksize;
read_ag_position = (xfs_off_t) read_ag_off * (xfs_off_t) BBSIZE;
read_ag_buf = malloc(read_ag_length);
if(read_ag_buf == NULL){
log_mesg(0, 1, 1, fs_opt.debug, "%s, %i, ERROR:%s", __func__, __LINE__, strerror(errno));
}
memset(read_ag_buf, 0, read_ag_length);
log_mesg(2, 0, 0, fs_opt.debug, "seek to read_ag_position %lli\n", read_ag_position);
sk = lseek(source_fd, read_ag_position, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (read_ag_length/source_blocksize);
set_bitmap(bitmap, sk, read_ag_length);
log_mesg(2, 0, 0, fs_opt.debug, "read ag header fd = %llu(%i), length = %i(%i)\n", sk, current_block, read_ag_length, block_count);
if ((res = read(source_fd, read_ag_buf, read_ag_length)) < 0) {
log_mesg(1, 0, 1, fs_opt.debug, "read failure at offset %lld\n", read_ag_position);
}
ag_hdr.xfs_sb = (xfs_dsb_t *) (read_ag_buf);
ASSERT(be32_to_cpu(ag_hdr.xfs_sb->sb_magicnum) == XFS_SB_MAGIC);
ag_hdr.xfs_agf = (xfs_agf_t *) (read_ag_buf + source_sectorsize);
ASSERT(be32_to_cpu(ag_hdr.xfs_agf->agf_magicnum) == XFS_AGF_MAGIC);
ag_hdr.xfs_agi = (xfs_agi_t *) (read_ag_buf + 2 * source_sectorsize);
ASSERT(be32_to_cpu(ag_hdr.xfs_agi->agi_magicnum) == XFS_AGI_MAGIC);
ag_hdr.xfs_agfl = (xfs_agfl_t *) (read_ag_buf + 3 * source_sectorsize);
log_mesg(2, 0, 0, fs_opt.debug, "ag header read ok\n");
/* save what we need (agf) in the btree buffer */
btree_buf_data = malloc(source_blocksize);
if(btree_buf_data == NULL){
log_mesg(0, 1, 1, fs_opt.debug, "%s, %i, ERROR:%s", __func__, __LINE__, strerror(errno));
}
memset(btree_buf_data, 0, source_blocksize);
memmove(btree_buf_data, ag_hdr.xfs_agf, source_sectorsize);
ag_hdr.xfs_agf = (xfs_agf_t *) btree_buf_data;
btree_buf_length = source_blocksize;
///* traverse btree until we get to the leftmost leaf node */
bno = be32_to_cpu(ag_hdr.xfs_agf->agf_roots[XFS_BTNUM_BNOi]);
current_level = 0;
btree_levels = be32_to_cpu(ag_hdr.xfs_agf->agf_levels[XFS_BTNUM_BNOi]);
ag_end = XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(ag_hdr.xfs_agf->agf_length) - 1) + source_blocksize / BBSIZE;
for (;;) {
/* none of this touches the w_buf buffer */
current_level++;
btree_buf_position = pos = (xfs_off_t)XFS_AGB_TO_DADDR(mp,agno,bno) << BBSHIFT;
btree_buf_length = source_blocksize;
sk = lseek(source_fd, btree_buf_position, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (btree_buf_length/source_blocksize);
set_bitmap(bitmap, sk, btree_buf_length);
log_mesg(2, 0, 0, fs_opt.debug, "read btree sf = %llu(%i), length = %i(%i)\n", sk, current_block, btree_buf_length, block_count);
read(source_fd, btree_buf_data, btree_buf_length);
block = (struct xfs_btree_block *)((char *)btree_buf_data + pos - btree_buf_position);
if (be16_to_cpu(block->bb_level) == 0)
break;
ptr = XFS_ALLOC_PTR_ADDR(mp, block, 1, mp->m_alloc_mxr[1]);
bno = be32_to_cpu(ptr[0]);
}
log_mesg(2, 0, 0, fs_opt.debug, "btree read done\n");
/* align first data copy but don't overwrite ag header */
pos = read_ag_position >> BBSHIFT;
length = read_ag_length >> BBSHIFT;
next_begin = pos + length;
ag_begin = next_begin;
///* handle the rest of the ag */
for (;;) {
if (be16_to_cpu(block->bb_level) != 0) {
log_mesg(0, 1, 1, fs_opt.debug, "WARNING: source filesystem inconsistent.\nA leaf btree rec isn't a leaf. Aborting now.\n");
}
rec_ptr = XFS_ALLOC_REC_ADDR(mp, block, 1);
for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++, rec_ptr++) {
/* calculate in daddr's */
begin = next_begin;
/*
* protect against pathological case of a
* hole right after the ag header in a
* mis-aligned case
*/
if (begin < ag_begin)
begin = ag_begin;
/*
* round size up to ensure we copy a
* range bigger than required
*/
log_mesg(3, 0, 0, fs_opt.debug, "XFS_AGB_TO_DADDR = %llu, agno = %i, be32_to_cpu=%llu\n", XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(rec_ptr->ar_startblock)), agno, be32_to_cpu(rec_ptr->ar_startblock));
sizeb = XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(rec_ptr->ar_startblock)) - begin;
size = roundup(sizeb <<BBSHIFT, source_sectorsize);
log_mesg(3, 0, 0, fs_opt.debug, "BB = %i size %i and sizeb %llu brgin = %llu\n", BBSHIFT, size, sizeb, begin);
if (size > 0) {
/* copy extent */
log_mesg(2, 0, 0, fs_opt.debug, "copy extent\n");
w_position = (xfs_off_t)begin << BBSHIFT;
while (size > 0) {
/*
* let lower layer do alignment
*/
if (size > w_size) {
w_length = w_size;
size -= w_size;
sizeb -= wblocks;
numblocks += wblocks;
} else {
w_length = size;
numblocks += sizeb;
size = 0;
}
//read_wbuf(source_fd, &w_buf, mp);
sk = lseek(source_fd, w_position, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (w_length/source_blocksize);
set_bitmap(bitmap, sk, w_length);
log_mesg(2, 0, 0, fs_opt.debug, "read ext sourcefd to w_buf source_fd=%llu(%i), length=%i(%i)\n", sk, current_block, w_length, block_count);
sk = lseek(source_fd, w_length, SEEK_CUR);
w_position += w_length;
}
}
/* round next starting point down */
new_begin = XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(rec_ptr->ar_startblock) + be32_to_cpu(rec_ptr->ar_blockcount));
next_begin = rounddown(new_begin, source_sectorsize >> BBSHIFT);
}
if (be32_to_cpu(block->bb_u.s.bb_rightsib) == NULLAGBLOCK){
log_mesg(2, 0, 0, fs_opt.debug, "NULLAGBLOCK\n");
break;
}
/* read in next btree record block */
btree_buf_position = pos = (xfs_off_t)XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(block->bb_u.s.bb_rightsib)) << BBSHIFT;
btree_buf_length = source_blocksize;
/* let read_wbuf handle alignment */
//read_wbuf(source_fd, &btree_buf, mp);
sk = lseek(source_fd, btree_buf_position, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (btree_buf_length/source_blocksize);
set_bitmap(bitmap, sk, btree_buf_length);
log_mesg(2, 0, 0, fs_opt.debug, "read btreebuf fd = %llu(%i), length = %i(%i) \n", sk, current_block, btree_buf_length, block_count);
read(source_fd, btree_buf_data, btree_buf_length);
block = (struct xfs_btree_block *)((char *) btree_buf_data + pos - btree_buf_position);
ASSERT(be32_to_cpu(block->bb_magic) == XFS_ABTB_MAGIC);
}
/*
* write out range of used blocks after last range
* of free blocks in AG
*/
if (next_begin < ag_end) {
begin = next_begin;
sizeb = ag_end - begin;
size = roundup(sizeb << BBSHIFT, source_sectorsize);
if (size > 0) {
/* copy extent */
w_position = (xfs_off_t) begin << BBSHIFT;
while (size > 0) {
/*
* let lower layer do alignment
*/
if (size > w_size) {
w_length = w_size;
size -= w_size;
sizeb -= wblocks;
numblocks += wblocks;
} else {
w_length = size;
numblocks += sizeb;
size = 0;
}
sk = lseek(source_fd, w_position, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (w_length/source_blocksize);
set_bitmap(bitmap, sk, w_length);
log_mesg(2, 0, 0, fs_opt.debug, "read ext fd = %llu(%i), length = %i(%i)\n", sk, current_block, w_length, block_count);
//read_wbuf(source_fd, &w_buf, mp);
lseek(source_fd, w_length, SEEK_CUR);
w_position += w_length;
}
}
}
log_mesg(2, 0, 0, fs_opt.debug, "write a clean log\n");
log_length = 1 * 1024 * 1024;
logstart = XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart) << BBSHIFT;
logstart_pos = rounddown(logstart, (xfs_off_t)log_length);
if (logstart % log_length) { /* unaligned */
sk = lseek(source_fd, logstart_pos, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (log_length/source_blocksize);
set_bitmap(bitmap, sk, log_length);
log_mesg(2, 0, 0, fs_opt.debug, "read log start from %llu(%i) %i(%i)\n", sk, current_block, log_length, block_count);
sk = lseek(source_fd, log_length, SEEK_CUR);
}
logend = XFS_FSB_TO_DADDR(mp, mp->m_sb.sb_logstart) << BBSHIFT;
logend += XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks);
logend_pos = rounddown(logend, (xfs_off_t)log_length);
if (logend % log_length) { /* unaligned */
sk = lseek(source_fd, logend_pos, SEEK_SET);
current_block = (sk/source_blocksize);
block_count = (log_length/source_blocksize);
set_bitmap(bitmap, sk, log_length);
log_mesg(2, 0, 0, fs_opt.debug, "read log end from %llu(%i) %i(%i)\n", sk, current_block, log_length, block_count);
sk = lseek(source_fd, log_length, SEEK_CUR);
}
log_mesg(2, 0, 0, fs_opt.debug, "write a clean log done\n");
prog_cur_block = image_hdr.totalblock/num_ags*(agno+1)-1;
update_pui(&prog, prog_cur_block, prog_cur_block, 0);
}
for(current_block = 0; current_block <= image_hdr.totalblock; current_block++){
if(pc_test_bit(current_block, bitmap))
bused++;
else
bfree++;
}
log_mesg(0, 0, 0, fs_opt.debug, "bused = %lli, bfree = %lli\n", bused, bfree);
fs_close();
update_pui(&prog, 1, 1, 1);
}
STATIC int
xfs_trim_extents(
struct xfs_mount *mp,
xfs_agnumber_t agno,
xfs_fsblock_t start,
xfs_fsblock_t len,
xfs_fsblock_t minlen,
__uint64_t *blocks_trimmed)
{
struct block_device *bdev = mp->m_ddev_targp->bt_bdev;
struct xfs_btree_cur *cur;
struct xfs_buf *agbp;
struct xfs_perag *pag;
int error;
int i;
pag = xfs_perag_get(mp, agno);
error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
if (error || !agbp)
goto out_put_perag;
cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
/*
* Force out the log. This means any transactions that might have freed
* space before we took the AGF buffer lock are now on disk, and the
* volatile disk cache is flushed.
*/
xfs_log_force(mp, XFS_LOG_SYNC);
/*
* Look up the longest btree in the AGF and start with it.
*/
error = xfs_alloc_lookup_le(cur, 0,
XFS_BUF_TO_AGF(agbp)->agf_longest, &i);
if (error)
goto out_del_cursor;
/*
* Loop until we are done with all extents that are large
* enough to be worth discarding.
*/
while (i) {
xfs_agblock_t fbno;
xfs_extlen_t flen;
error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
if (error)
goto out_del_cursor;
XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest);
/*
* Too small? Give up.
*/
if (flen < minlen) {
trace_xfs_discard_toosmall(mp, agno, fbno, flen);
goto out_del_cursor;
}
/*
* If the extent is entirely outside of the range we are
* supposed to discard skip it. Do not bother to trim
* down partially overlapping ranges for now.
*/
if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start ||
XFS_AGB_TO_FSB(mp, agno, fbno) >= start + len) {
trace_xfs_discard_exclude(mp, agno, fbno, flen);
goto next_extent;
}
/*
* If any blocks in the range are still busy, skip the
* discard and try again the next time.
*/
if (xfs_alloc_busy_search(mp, agno, fbno, flen)) {
trace_xfs_discard_busy(mp, agno, fbno, flen);
goto next_extent;
}
trace_xfs_discard_extent(mp, agno, fbno, flen);
error = -blkdev_issue_discard(bdev,
XFS_AGB_TO_DADDR(mp, agno, fbno),
XFS_FSB_TO_BB(mp, flen),
GFP_NOFS, 0);
if (error)
goto out_del_cursor;
*blocks_trimmed += flen;
next_extent:
error = xfs_btree_decrement(cur, 0, &i);
if (error)
goto out_del_cursor;
}
out_del_cursor:
xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
xfs_buf_relse(agbp);
out_put_perag:
xfs_perag_put(pag);
return error;
}
/*
* rebuilds an inode tree given a cursor. We're lazy here and call
* the routine that builds the agi
*/
static void
build_ino_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
bt_status_t *btree_curs, __uint32_t magic,
struct agi_stat *agi_stat, int finobt)
{
xfs_agnumber_t i;
xfs_agblock_t j;
xfs_agblock_t agbno;
xfs_agino_t first_agino;
struct xfs_btree_block *bt_hdr;
xfs_inobt_rec_t *bt_rec;
ino_tree_node_t *ino_rec;
bt_stat_level_t *lptr;
xfs_agino_t count = 0;
xfs_agino_t freecount = 0;
int inocnt;
uint8_t finocnt;
int k;
int level = btree_curs->num_levels;
int spmask;
uint64_t sparse;
uint16_t holemask;
for (i = 0; i < level; i++) {
lptr = &btree_curs->level[i];
agbno = get_next_blockaddr(agno, i, btree_curs);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
XFS_FSB_TO_BB(mp, 1));
if (i == btree_curs->num_levels - 1)
btree_curs->root = agbno;
lptr->agbno = agbno;
lptr->prev_agbno = NULLAGBLOCK;
lptr->prev_buf_p = NULL;
/*
* initialize block header
*/
lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, magic,
i, 0, agno,
XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, magic,
i, 0, agno, 0);
}
/*
* run along leaf, setting up records. as we have to switch
* blocks, call the prop_ino_cursor routine to set up the new
* pointers for the parent. that can recurse up to the root
* if required. set the sibling pointers for leaf level here.
*/
if (finobt)
ino_rec = findfirst_free_inode_rec(agno);
else
ino_rec = findfirst_inode_rec(agno);
if (ino_rec != NULL)
first_agino = ino_rec->ino_startnum;
else
first_agino = NULLAGINO;
lptr = &btree_curs->level[0];
for (i = 0; i < lptr->num_blocks; i++) {
/*
* block initialization, lay in block header
*/
lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, magic,
0, 0, agno,
XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, magic,
0, 0, agno, 0);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
bt_hdr->bb_numrecs = cpu_to_be16(lptr->num_recs_pb +
(lptr->modulo > 0));
if (lptr->modulo > 0)
lptr->modulo--;
if (lptr->num_recs_pb > 0)
prop_ino_cursor(mp, agno, btree_curs,
ino_rec->ino_startnum, 0);
bt_rec = (xfs_inobt_rec_t *)
((char *)bt_hdr + XFS_INOBT_BLOCK_LEN(mp));
for (j = 0; j < be16_to_cpu(bt_hdr->bb_numrecs); j++) {
ASSERT(ino_rec != NULL);
bt_rec[j].ir_startino =
cpu_to_be32(ino_rec->ino_startnum);
bt_rec[j].ir_free = cpu_to_be64(ino_rec->ir_free);
inocnt = finocnt = 0;
for (k = 0; k < sizeof(xfs_inofree_t)*NBBY; k++) {
ASSERT(is_inode_confirmed(ino_rec, k));
if (is_inode_sparse(ino_rec, k))
continue;
if (is_inode_free(ino_rec, k))
finocnt++;
inocnt++;
}
/*
* Set the freecount and check whether we need to update
* the sparse format fields. Otherwise, skip to the next
* record.
*/
inorec_set_freecount(mp, &bt_rec[j], finocnt);
if (!xfs_sb_version_hassparseinodes(&mp->m_sb))
goto nextrec;
/*
* Convert the 64-bit in-core sparse inode state to the
* 16-bit on-disk holemask.
*/
holemask = 0;
spmask = (1 << XFS_INODES_PER_HOLEMASK_BIT) - 1;
sparse = ino_rec->ir_sparse;
for (k = 0; k < XFS_INOBT_HOLEMASK_BITS; k++) {
if (sparse & spmask) {
ASSERT((sparse & spmask) == spmask);
holemask |= (1 << k);
} else
ASSERT((sparse & spmask) == 0);
sparse >>= XFS_INODES_PER_HOLEMASK_BIT;
}
bt_rec[j].ir_u.sp.ir_count = inocnt;
bt_rec[j].ir_u.sp.ir_holemask = cpu_to_be16(holemask);
nextrec:
freecount += finocnt;
count += inocnt;
if (finobt)
ino_rec = next_free_ino_rec(ino_rec);
else
ino_rec = next_ino_rec(ino_rec);
}
if (ino_rec != NULL) {
/*
* get next leaf level block
*/
if (lptr->prev_buf_p != NULL) {
#ifdef XR_BLD_INO_TRACE
fprintf(stderr, "writing inobt agbno %u\n",
lptr->prev_agbno);
#endif
ASSERT(lptr->prev_agbno != NULLAGBLOCK);
libxfs_writebuf(lptr->prev_buf_p, 0);
}
lptr->prev_buf_p = lptr->buf_p;
lptr->prev_agbno = lptr->agbno;
lptr->agbno = get_next_blockaddr(agno, 0, btree_curs);
bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(lptr->agbno);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, lptr->agbno),
XFS_FSB_TO_BB(mp, 1));
}
}
if (agi_stat) {
agi_stat->first_agino = first_agino;
agi_stat->count = count;
agi_stat->freecount = freecount;
}
}
int
main(int argc, char **argv)
{
int i, j;
int howfar = 0;
int open_flags;
xfs_off_t pos, end_pos;
size_t length;
int c, first_residue, tmp_residue;
__uint64_t size, sizeb;
__uint64_t numblocks = 0;
int wblocks = 0;
int num_threads = 0;
struct dioattr d;
int wbuf_size;
int wbuf_align;
int wbuf_miniosize;
int source_is_file = 0;
int buffered_output = 0;
int duplicate = 0;
uint btree_levels, current_level;
ag_header_t ag_hdr;
xfs_mount_t *mp;
xfs_mount_t mbuf;
xfs_buf_t *sbp;
xfs_sb_t *sb;
xfs_agnumber_t num_ags, agno;
xfs_agblock_t bno;
xfs_daddr_t begin, next_begin, ag_begin, new_begin, ag_end;
struct xfs_btree_block *block;
xfs_alloc_ptr_t *ptr;
xfs_alloc_rec_t *rec_ptr;
extern char *optarg;
extern int optind;
libxfs_init_t xargs;
thread_args *tcarg;
struct stat64 statbuf;
progname = basename(argv[0]);
setlocale(LC_ALL, "");
bindtextdomain(PACKAGE, LOCALEDIR);
textdomain(PACKAGE);
while ((c = getopt(argc, argv, "bdL:V")) != EOF) {
switch (c) {
case 'b':
buffered_output = 1;
break;
case 'd':
duplicate = 1;
break;
case 'L':
logfile_name = optarg;
break;
case 'V':
printf(_("%s version %s\n"), progname, VERSION);
exit(0);
case '?':
usage();
}
}
if (argc - optind < 2)
usage();
if (logfile_name) {
logfd = open(logfile_name, O_CREAT|O_WRONLY|O_EXCL, 0600);
} else {
logfile_name = LOGFILE_NAME;
logfd = mkstemp(logfile_name);
}
if (logfd < 0) {
fprintf(stderr, _("%s: couldn't open log file \"%s\"\n"),
progname, logfile_name);
perror(_("Aborting XFS copy - reason"));
exit(1);
}
if ((logerr = fdopen(logfd, "w")) == NULL) {
fprintf(stderr, _("%s: couldn't set up logfile stream\n"),
progname);
perror(_("Aborting XFS copy - reason"));
exit(1);
}
source_name = argv[optind];
source_fd = -1;
optind++;
num_targets = argc - optind;
if ((target = malloc(sizeof(target_control) * num_targets)) == NULL) {
do_log(_("Couldn't allocate target array\n"));
die_perror();
}
for (i = 0; optind < argc; i++, optind++) {
target[i].name = argv[optind];
target[i].fd = -1;
target[i].position = -1;
target[i].state = INACTIVE;
target[i].error = 0;
target[i].err_type = 0;
}
parent_pid = getpid();
if (atexit(killall)) {
do_log(_("%s: couldn't register atexit function.\n"), progname);
die_perror();
}
/* open up source -- is it a file? */
open_flags = O_RDONLY;
if ((source_fd = open(source_name, open_flags)) < 0) {
do_log(_("%s: couldn't open source \"%s\"\n"),
progname, source_name);
die_perror();
}
if (fstat64(source_fd, &statbuf) < 0) {
do_log(_("%s: couldn't stat source \"%s\"\n"),
progname, source_name);
die_perror();
}
if (S_ISREG(statbuf.st_mode))
source_is_file = 1;
if (source_is_file && platform_test_xfs_fd(source_fd)) {
if (fcntl(source_fd, F_SETFL, open_flags | O_DIRECT) < 0) {
do_log(_("%s: Cannot set direct I/O flag on \"%s\".\n"),
progname, source_name);
die_perror();
}
if (xfsctl(source_name, source_fd, XFS_IOC_DIOINFO, &d) < 0) {
do_log(_("%s: xfsctl on file \"%s\" failed.\n"),
progname, source_name);
die_perror();
}
wbuf_align = d.d_mem;
wbuf_size = MIN(d.d_maxiosz, 1 * 1024 * 1024);
wbuf_miniosize = d.d_miniosz;
} else {
/* set arbitrary I/O params, miniosize at least 1 disk block */
wbuf_align = getpagesize();
wbuf_size = 1 * 1024 * 1024;
wbuf_miniosize = -1; /* set after mounting source fs */
}
if (!source_is_file) {
/*
* check to make sure a filesystem isn't mounted
* on the device
*/
if (platform_check_ismounted(source_name, NULL, &statbuf, 0)) {
do_log(
_("%s: Warning -- a filesystem is mounted on the source device.\n"),
progname);
do_log(
_("\t\tGenerated copies may be corrupt unless the source is\n"));
do_log(
_("\t\tunmounted or mounted read-only. Copy proceeding...\n"));
}
}
/* prepare the libxfs_init structure */
memset(&xargs, 0, sizeof(xargs));
xargs.isdirect = LIBXFS_DIRECT;
xargs.isreadonly = LIBXFS_ISREADONLY;
if (source_is_file) {
xargs.dname = source_name;
xargs.disfile = 1;
} else
xargs.volname = source_name;
if (!libxfs_init(&xargs)) {
do_log(_("%s: couldn't initialize XFS library\n"
"%s: Aborting.\n"), progname, progname);
exit(1);
}
/* prepare the mount structure */
sbp = libxfs_readbuf(xargs.ddev, XFS_SB_DADDR, 1, 0);
memset(&mbuf, 0, sizeof(xfs_mount_t));
sb = &mbuf.m_sb;
libxfs_sb_from_disk(sb, XFS_BUF_TO_SBP(sbp));
mp = libxfs_mount(&mbuf, sb, xargs.ddev, xargs.logdev, xargs.rtdev, 1);
if (mp == NULL) {
do_log(_("%s: %s filesystem failed to initialize\n"
"%s: Aborting.\n"), progname, source_name, progname);
exit(1);
} else if (mp->m_sb.sb_inprogress) {
do_log(_("%s %s filesystem failed to initialize\n"
"%s: Aborting.\n"), progname, source_name, progname);
exit(1);
} else if (mp->m_sb.sb_logstart == 0) {
do_log(_("%s: %s has an external log.\n%s: Aborting.\n"),
progname, source_name, progname);
exit(1);
} else if (mp->m_sb.sb_rextents != 0) {
do_log(_("%s: %s has a real-time section.\n"
"%s: Aborting.\n"), progname, source_name, progname);
exit(1);
}
source_blocksize = mp->m_sb.sb_blocksize;
source_sectorsize = mp->m_sb.sb_sectsize;
if (wbuf_miniosize == -1)
wbuf_miniosize = source_sectorsize;
ASSERT(source_blocksize % source_sectorsize == 0);
ASSERT(source_sectorsize % BBSIZE == 0);
if (source_blocksize > source_sectorsize) {
/* get number of leftover sectors in last block of ag header */
tmp_residue = ((XFS_AGFL_DADDR(mp) + 1) * source_sectorsize)
% source_blocksize;
first_residue = (tmp_residue == 0) ? 0 :
source_blocksize - tmp_residue;
ASSERT(first_residue % source_sectorsize == 0);
} else if (source_blocksize == source_sectorsize) {
first_residue = 0;
} else {
do_log(_("Error: filesystem block size is smaller than the"
" disk sectorsize.\nAborting XFS copy now.\n"));
exit(1);
}
first_agbno = (((XFS_AGFL_DADDR(mp) + 1) * source_sectorsize)
+ first_residue) / source_blocksize;
ASSERT(first_agbno != 0);
ASSERT( ((((XFS_AGFL_DADDR(mp) + 1) * source_sectorsize)
+ first_residue) % source_blocksize) == 0);
/* now open targets */
open_flags = O_RDWR;
for (i = 0; i < num_targets; i++) {
int write_last_block = 0;
if (stat64(target[i].name, &statbuf) < 0) {
/* ok, assume it's a file and create it */
do_out(_("Creating file %s\n"), target[i].name);
open_flags |= O_CREAT;
if (!buffered_output)
open_flags |= O_DIRECT;
write_last_block = 1;
} else if (S_ISREG(statbuf.st_mode)) {
open_flags |= O_TRUNC;
if (!buffered_output)
open_flags |= O_DIRECT;
write_last_block = 1;
} else {
/*
* check to make sure a filesystem isn't mounted
* on the device
*/
if (platform_check_ismounted(target[i].name,
NULL, &statbuf, 0)) {
do_log(_("%s: a filesystem is mounted "
"on target device \"%s\".\n"
"%s cannot copy to mounted filesystems."
" Aborting\n"),
progname, target[i].name, progname);
exit(1);
}
}
target[i].fd = open(target[i].name, open_flags, 0644);
if (target[i].fd < 0) {
do_log(_("%s: couldn't open target \"%s\"\n"),
progname, target[i].name);
die_perror();
}
if (write_last_block) {
/* ensure regular files are correctly sized */
if (ftruncate64(target[i].fd, mp->m_sb.sb_dblocks *
source_blocksize)) {
do_log(_("%s: cannot grow data section.\n"),
progname);
die_perror();
}
if (platform_test_xfs_fd(target[i].fd)) {
if (xfsctl(target[i].name, target[i].fd,
XFS_IOC_DIOINFO, &d) < 0) {
do_log(
_("%s: xfsctl on \"%s\" failed.\n"),
progname, target[i].name);
die_perror();
} else {
wbuf_align = MAX(wbuf_align, d.d_mem);
wbuf_size = MIN(d.d_maxiosz, wbuf_size);
wbuf_miniosize = MAX(d.d_miniosz,
wbuf_miniosize);
}
}
} else {
char *lb[XFS_MAX_SECTORSIZE] = { NULL };
off64_t off;
/* ensure device files are sufficiently large */
off = mp->m_sb.sb_dblocks * source_blocksize;
off -= sizeof(lb);
if (pwrite64(target[i].fd, lb, sizeof(lb), off) < 0) {
do_log(_("%s: failed to write last block\n"),
progname);
do_log(_("\tIs target \"%s\" too small?\n"),
target[i].name);
die_perror();
}
}
}
/* initialize locks and bufs */
if (pthread_mutex_init(&glob_masks.mutex, NULL) != 0) {
do_log(_("Couldn't initialize global thread mask\n"));
die_perror();
}
glob_masks.num_working = 0;
if (wbuf_init(&w_buf, wbuf_size, wbuf_align,
wbuf_miniosize, 0) == NULL) {
do_log(_("Error initializing wbuf 0\n"));
die_perror();
}
wblocks = wbuf_size / BBSIZE;
if (wbuf_init(&btree_buf, MAX(source_blocksize, wbuf_miniosize),
wbuf_align, wbuf_miniosize, 1) == NULL) {
do_log(_("Error initializing btree buf 1\n"));
die_perror();
}
if (pthread_mutex_init(&mainwait,NULL) != 0) {
do_log(_("Error creating first semaphore.\n"));
die_perror();
exit(1);
}
/* need to start out blocking */
pthread_mutex_lock(&mainwait);
/* set up sigchild signal handler */
signal(SIGCHLD, handler);
signal_maskfunc(SIGCHLD, SIG_BLOCK);
/* make children */
if ((targ = malloc(num_targets * sizeof(thread_args))) == NULL) {
do_log(_("Couldn't malloc space for thread args\n"));
die_perror();
exit(1);
}
for (i = 0, tcarg = targ; i < num_targets; i++, tcarg++) {
if (!duplicate)
platform_uuid_generate(&tcarg->uuid);
else
platform_uuid_copy(&tcarg->uuid, &mp->m_sb.sb_uuid);
if (pthread_mutex_init(&tcarg->wait, NULL) != 0) {
do_log(_("Error creating thread mutex %d\n"), i);
die_perror();
exit(1);
}
/* need to start out blocking */
pthread_mutex_lock(&tcarg->wait);
}
for (i = 0, tcarg = targ; i < num_targets; i++, tcarg++) {
tcarg->id = i;
tcarg->fd = target[i].fd;
target[i].state = ACTIVE;
num_threads++;
if (pthread_create(&target[i].pid, NULL,
begin_reader, (void *)tcarg)) {
do_log(_("Error creating thread for target %d\n"), i);
die_perror();
}
}
ASSERT(num_targets == num_threads);
/* set up statistics */
num_ags = mp->m_sb.sb_agcount;
init_bar(mp->m_sb.sb_blocksize / BBSIZE
* ((__uint64_t)mp->m_sb.sb_dblocks
- (__uint64_t)mp->m_sb.sb_fdblocks + 10 * num_ags));
kids = num_targets;
block = (struct xfs_btree_block *) btree_buf.data;
for (agno = 0; agno < num_ags && kids > 0; agno++) {
/* read in first blocks of the ag */
read_ag_header(source_fd, agno, &w_buf, &ag_hdr, mp,
source_blocksize, source_sectorsize);
/* set the in_progress bit for the first AG */
if (agno == 0)
ag_hdr.xfs_sb->sb_inprogress = 1;
/* save what we need (agf) in the btree buffer */
memmove(btree_buf.data, ag_hdr.xfs_agf, source_sectorsize);
ag_hdr.xfs_agf = (xfs_agf_t *) btree_buf.data;
btree_buf.length = source_blocksize;
/* write the ag header out */
write_wbuf();
/* traverse btree until we get to the leftmost leaf node */
bno = be32_to_cpu(ag_hdr.xfs_agf->agf_roots[XFS_BTNUM_BNOi]);
current_level = 0;
btree_levels = be32_to_cpu(ag_hdr.xfs_agf->
agf_levels[XFS_BTNUM_BNOi]);
ag_end = XFS_AGB_TO_DADDR(mp, agno,
be32_to_cpu(ag_hdr.xfs_agf->agf_length) - 1)
+ source_blocksize / BBSIZE;
for (;;) {
/* none of this touches the w_buf buffer */
ASSERT(current_level < btree_levels);
current_level++;
btree_buf.position = pos = (xfs_off_t)
XFS_AGB_TO_DADDR(mp,agno,bno) << BBSHIFT;
btree_buf.length = source_blocksize;
read_wbuf(source_fd, &btree_buf, mp);
block = (struct xfs_btree_block *)
((char *)btree_buf.data +
pos - btree_buf.position);
ASSERT(be32_to_cpu(block->bb_magic) == XFS_ABTB_MAGIC);
if (be16_to_cpu(block->bb_level) == 0)
break;
ptr = XFS_ALLOC_PTR_ADDR(mp, block, 1,
mp->m_alloc_mxr[1]);
bno = be32_to_cpu(ptr[0]);
}
/* align first data copy but don't overwrite ag header */
pos = w_buf.position >> BBSHIFT;
length = w_buf.length >> BBSHIFT;
next_begin = pos + length;
ag_begin = next_begin;
ASSERT(w_buf.position % source_sectorsize == 0);
/* handle the rest of the ag */
for (;;) {
if (be16_to_cpu(block->bb_level) != 0) {
do_log(
_("WARNING: source filesystem inconsistent.\n"));
do_log(
_(" A leaf btree rec isn't a leaf. Aborting now.\n"));
exit(1);
}
rec_ptr = XFS_ALLOC_REC_ADDR(mp, block, 1);
for (i = 0; i < be16_to_cpu(block->bb_numrecs);
i++, rec_ptr++) {
/* calculate in daddr's */
begin = next_begin;
/*
* protect against pathological case of a
* hole right after the ag header in a
* mis-aligned case
*/
if (begin < ag_begin)
begin = ag_begin;
/*
* round size up to ensure we copy a
* range bigger than required
*/
sizeb = XFS_AGB_TO_DADDR(mp, agno,
be32_to_cpu(rec_ptr->ar_startblock)) -
begin;
size = roundup(sizeb <<BBSHIFT, wbuf_miniosize);
if (size > 0) {
/* copy extent */
w_buf.position = (xfs_off_t)
begin << BBSHIFT;
while (size > 0) {
/*
* let lower layer do alignment
*/
if (size > w_buf.size) {
w_buf.length = w_buf.size;
size -= w_buf.size;
sizeb -= wblocks;
numblocks += wblocks;
} else {
w_buf.length = size;
numblocks += sizeb;
size = 0;
}
read_wbuf(source_fd, &w_buf, mp);
write_wbuf();
w_buf.position += w_buf.length;
howfar = bump_bar(
howfar, numblocks);
}
}
/* round next starting point down */
new_begin = XFS_AGB_TO_DADDR(mp, agno,
be32_to_cpu(rec_ptr->ar_startblock) +
be32_to_cpu(rec_ptr->ar_blockcount));
next_begin = rounddown(new_begin,
w_buf.min_io_size >> BBSHIFT);
}
if (be32_to_cpu(block->bb_u.s.bb_rightsib) == NULLAGBLOCK)
break;
/* read in next btree record block */
btree_buf.position = pos = (xfs_off_t)
XFS_AGB_TO_DADDR(mp, agno, be32_to_cpu(
block->bb_u.s.bb_rightsib)) << BBSHIFT;
btree_buf.length = source_blocksize;
/* let read_wbuf handle alignment */
read_wbuf(source_fd, &btree_buf, mp);
block = (struct xfs_btree_block *)
((char *) btree_buf.data +
pos - btree_buf.position);
ASSERT(be32_to_cpu(block->bb_magic) == XFS_ABTB_MAGIC);
}
/*
* write out range of used blocks after last range
* of free blocks in AG
*/
if (next_begin < ag_end) {
begin = next_begin;
sizeb = ag_end - begin;
size = roundup(sizeb << BBSHIFT, wbuf_miniosize);
if (size > 0) {
/* copy extent */
w_buf.position = (xfs_off_t) begin << BBSHIFT;
while (size > 0) {
/*
* let lower layer do alignment
*/
if (size > w_buf.size) {
w_buf.length = w_buf.size;
size -= w_buf.size;
sizeb -= wblocks;
numblocks += wblocks;
} else {
w_buf.length = size;
numblocks += sizeb;
size = 0;
}
read_wbuf(source_fd, &w_buf, mp);
write_wbuf();
w_buf.position += w_buf.length;
howfar = bump_bar(howfar, numblocks);
}
}
}
}
if (kids > 0) {
if (!duplicate) {
/* write a clean log using the specified UUID */
for (j = 0, tcarg = targ; j < num_targets; j++) {
w_buf.owner = tcarg;
w_buf.length = rounddown(w_buf.size,
w_buf.min_io_size);
pos = write_log_header(
source_fd, &w_buf, mp);
end_pos = write_log_trailer(
source_fd, &w_buf, mp);
w_buf.position = pos;
memset(w_buf.data, 0, w_buf.length);
while (w_buf.position < end_pos) {
do_write(tcarg);
w_buf.position += w_buf.length;
}
tcarg++;
}
} else {
num_ags = 1;
}
/* reread and rewrite superblocks (UUID and in-progress) */
/* [backwards, so inprogress bit only updated when done] */
for (i = num_ags - 1; i >= 0; i--) {
read_ag_header(source_fd, i, &w_buf, &ag_hdr, mp,
source_blocksize, source_sectorsize);
if (i == 0)
ag_hdr.xfs_sb->sb_inprogress = 0;
/* do each thread in turn, each has its own UUID */
for (j = 0, tcarg = targ; j < num_targets; j++) {
platform_uuid_copy(&ag_hdr.xfs_sb->sb_uuid,
&tcarg->uuid);
do_write(tcarg);
tcarg++;
}
}
bump_bar(100, 0);
}
check_errors();
killall();
pthread_exit(NULL);
/*NOTREACHED*/
return 0;
}
/* Does this block match the btree information passed in? */
STATIC int
xrep_findroot_block(
struct xrep_findroot *ri,
struct xrep_find_ag_btree *fab,
uint64_t owner,
xfs_agblock_t agbno,
bool *done_with_block)
{
struct xfs_mount *mp = ri->sc->mp;
struct xfs_buf *bp;
struct xfs_btree_block *btblock;
xfs_daddr_t daddr;
int block_level;
int error = 0;
daddr = XFS_AGB_TO_DADDR(mp, ri->sc->sa.agno, agbno);
/*
* Blocks in the AGFL have stale contents that might just happen to
* have a matching magic and uuid. We don't want to pull these blocks
* in as part of a tree root, so we have to filter out the AGFL stuff
* here. If the AGFL looks insane we'll just refuse to repair.
*/
if (owner == XFS_RMAP_OWN_AG) {
error = xfs_agfl_walk(mp, ri->agf, ri->agfl_bp,
xrep_findroot_agfl_walk, &agbno);
if (error == XFS_BTREE_QUERY_RANGE_ABORT)
return 0;
if (error)
return error;
}
/*
* Read the buffer into memory so that we can see if it's a match for
* our btree type. We have no clue if it is beforehand, and we want to
* avoid xfs_trans_read_buf's behavior of dumping the DONE state (which
* will cause needless disk reads in subsequent calls to this function)
* and logging metadata verifier failures.
*
* Therefore, pass in NULL buffer ops. If the buffer was already in
* memory from some other caller it will already have b_ops assigned.
* If it was in memory from a previous unsuccessful findroot_block
* call, the buffer won't have b_ops but it should be clean and ready
* for us to try to verify if the read call succeeds. The same applies
* if the buffer wasn't in memory at all.
*
* Note: If we never match a btree type with this buffer, it will be
* left in memory with NULL b_ops. This shouldn't be a problem unless
* the buffer gets written.
*/
error = xfs_trans_read_buf(mp, ri->sc->tp, mp->m_ddev_targp, daddr,
mp->m_bsize, 0, &bp, NULL);
if (error)
return error;
/* Ensure the block magic matches the btree type we're looking for. */
btblock = XFS_BUF_TO_BLOCK(bp);
ASSERT(fab->buf_ops->magic[1] != 0);
if (btblock->bb_magic != fab->buf_ops->magic[1])
goto out;
/*
* If the buffer already has ops applied and they're not the ones for
* this btree type, we know this block doesn't match the btree and we
* can bail out.
*
* If the buffer ops match ours, someone else has already validated
* the block for us, so we can move on to checking if this is a root
* block candidate.
*
* If the buffer does not have ops, nobody has successfully validated
* the contents and the buffer cannot be dirty. If the magic, uuid,
* and structure match this btree type then we'll move on to checking
* if it's a root block candidate. If there is no match, bail out.
*/
if (bp->b_ops) {
if (bp->b_ops != fab->buf_ops)
goto out;
} else {
ASSERT(!xfs_trans_buf_is_dirty(bp));
if (!uuid_equal(&btblock->bb_u.s.bb_uuid,
&mp->m_sb.sb_meta_uuid))
goto out;
/*
* Read verifiers can reference b_ops, so we set the pointer
* here. If the verifier fails we'll reset the buffer state
* to what it was before we touched the buffer.
*/
bp->b_ops = fab->buf_ops;
fab->buf_ops->verify_read(bp);
if (bp->b_error) {
bp->b_ops = NULL;
bp->b_error = 0;
goto out;
}
/*
* Some read verifiers will (re)set b_ops, so we must be
* careful not to change b_ops after running the verifier.
*/
}
/*
* This block passes the magic/uuid and verifier tests for this btree
* type. We don't need the caller to try the other tree types.
*/
*done_with_block = true;
/*
* Compare this btree block's level to the height of the current
* candidate root block.
*
* If the level matches the root we found previously, throw away both
* blocks because there can't be two candidate roots.
*
* If level is lower in the tree than the root we found previously,
* ignore this block.
*/
block_level = xfs_btree_get_level(btblock);
if (block_level + 1 == fab->height) {
fab->root = NULLAGBLOCK;
goto out;
} else if (block_level < fab->height) {
goto out;
}
/*
* This is the highest block in the tree that we've found so far.
* Update the btree height to reflect what we've learned from this
* block.
*/
fab->height = block_level + 1;
/*
* If this block doesn't have sibling pointers, then it's the new root
* block candidate. Otherwise, the root will be found farther up the
* tree.
*/
if (btblock->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK) &&
btblock->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK))
fab->root = agbno;
else
fab->root = NULLAGBLOCK;
trace_xrep_findroot_block(mp, ri->sc->sa.agno, agbno,
be32_to_cpu(btblock->bb_magic), fab->height - 1);
out:
xfs_trans_brelse(ri->sc->tp, bp);
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;
}
/*
* rebuilds a freespace tree given a cursor and magic number of type
* of tree to build (bno or bcnt). returns the number of free blocks
* represented by the tree.
*/
static xfs_extlen_t
build_freespace_tree(xfs_mount_t *mp, xfs_agnumber_t agno,
bt_status_t *btree_curs, __uint32_t magic)
{
xfs_agnumber_t i;
xfs_agblock_t j;
struct xfs_btree_block *bt_hdr;
xfs_alloc_rec_t *bt_rec;
int level;
xfs_agblock_t agbno;
extent_tree_node_t *ext_ptr;
bt_stat_level_t *lptr;
xfs_extlen_t freeblks;
__uint32_t crc_magic;
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, "in build_freespace_tree, agno = %d\n", agno);
#endif
level = btree_curs->num_levels;
freeblks = 0;
ASSERT(level > 0);
if (magic == XFS_ABTB_MAGIC)
crc_magic = XFS_ABTB_CRC_MAGIC;
else
crc_magic = XFS_ABTC_CRC_MAGIC;
/*
* initialize the first block on each btree level
*/
for (i = 0; i < level; i++) {
lptr = &btree_curs->level[i];
agbno = get_next_blockaddr(agno, i, btree_curs);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
XFS_FSB_TO_BB(mp, 1));
if (i == btree_curs->num_levels - 1)
btree_curs->root = agbno;
lptr->agbno = agbno;
lptr->prev_agbno = NULLAGBLOCK;
lptr->prev_buf_p = NULL;
/*
* initialize block header
*/
lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, crc_magic, i,
0, agno, XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, magic, i,
0, agno, 0);
}
/*
* run along leaf, setting up records. as we have to switch
* blocks, call the prop_freespace_cursor routine to set up the new
* pointers for the parent. that can recurse up to the root
* if required. set the sibling pointers for leaf level here.
*/
if (magic == XFS_ABTB_MAGIC)
ext_ptr = findfirst_bno_extent(agno);
else
ext_ptr = findfirst_bcnt_extent(agno);
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, "bft, agno = %d, start = %u, count = %u\n",
agno, ext_ptr->ex_startblock, ext_ptr->ex_blockcount);
#endif
lptr = &btree_curs->level[0];
for (i = 0; i < btree_curs->level[0].num_blocks; i++) {
/*
* block initialization, lay in block header
*/
lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, crc_magic, 0,
0, agno, XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, magic, 0,
0, agno, 0);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
bt_hdr->bb_numrecs = cpu_to_be16(lptr->num_recs_pb +
(lptr->modulo > 0));
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, "bft, bb_numrecs = %d\n",
be16_to_cpu(bt_hdr->bb_numrecs));
#endif
if (lptr->modulo > 0)
lptr->modulo--;
/*
* initialize values in the path up to the root if
* this is a multi-level btree
*/
if (btree_curs->num_levels > 1)
prop_freespace_cursor(mp, agno, btree_curs,
ext_ptr->ex_startblock,
ext_ptr->ex_blockcount,
0, magic);
bt_rec = (xfs_alloc_rec_t *)
((char *)bt_hdr + XFS_ALLOC_BLOCK_LEN(mp));
for (j = 0; j < be16_to_cpu(bt_hdr->bb_numrecs); j++) {
ASSERT(ext_ptr != NULL);
bt_rec[j].ar_startblock = cpu_to_be32(
ext_ptr->ex_startblock);
bt_rec[j].ar_blockcount = cpu_to_be32(
ext_ptr->ex_blockcount);
freeblks += ext_ptr->ex_blockcount;
if (magic == XFS_ABTB_MAGIC)
ext_ptr = findnext_bno_extent(ext_ptr);
else
ext_ptr = findnext_bcnt_extent(agno, ext_ptr);
#if 0
#ifdef XR_BLD_FREE_TRACE
if (ext_ptr == NULL)
fprintf(stderr, "null extent pointer, j = %d\n",
j);
else
fprintf(stderr,
"bft, agno = %d, start = %u, count = %u\n",
agno, ext_ptr->ex_startblock,
ext_ptr->ex_blockcount);
#endif
#endif
}
if (ext_ptr != NULL) {
/*
* get next leaf level block
*/
if (lptr->prev_buf_p != NULL) {
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, " writing fst agbno %u\n",
lptr->prev_agbno);
#endif
ASSERT(lptr->prev_agbno != NULLAGBLOCK);
libxfs_writebuf(lptr->prev_buf_p, 0);
}
lptr->prev_buf_p = lptr->buf_p;
lptr->prev_agbno = lptr->agbno;
lptr->agbno = get_next_blockaddr(agno, 0, btree_curs);
bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(lptr->agbno);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, lptr->agbno),
XFS_FSB_TO_BB(mp, 1));
}
}
return(freeblks);
}
static void
prop_freespace_cursor(xfs_mount_t *mp, xfs_agnumber_t agno,
bt_status_t *btree_curs, xfs_agblock_t startblock,
xfs_extlen_t blockcount, int level, __uint32_t magic)
{
struct xfs_btree_block *bt_hdr;
xfs_alloc_key_t *bt_key;
xfs_alloc_ptr_t *bt_ptr;
xfs_agblock_t agbno;
bt_stat_level_t *lptr;
__uint32_t crc_magic;
if (magic == XFS_ABTB_MAGIC)
crc_magic = XFS_ABTB_CRC_MAGIC;
else
crc_magic = XFS_ABTC_CRC_MAGIC;
level++;
if (level >= btree_curs->num_levels)
return;
lptr = &btree_curs->level[level];
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
if (be16_to_cpu(bt_hdr->bb_numrecs) == 0) {
/*
* only happens once when initializing the
* left-hand side of the tree.
*/
prop_freespace_cursor(mp, agno, btree_curs, startblock,
blockcount, level, magic);
}
if (be16_to_cpu(bt_hdr->bb_numrecs) ==
lptr->num_recs_pb + (lptr->modulo > 0)) {
/*
* write out current prev block, grab us a new block,
* and set the rightsib pointer of current block
*/
#ifdef XR_BLD_FREE_TRACE
fprintf(stderr, " %d ", lptr->prev_agbno);
#endif
if (lptr->prev_agbno != NULLAGBLOCK) {
ASSERT(lptr->prev_buf_p != NULL);
libxfs_writebuf(lptr->prev_buf_p, 0);
}
lptr->prev_agbno = lptr->agbno;;
lptr->prev_buf_p = lptr->buf_p;
agbno = get_next_blockaddr(agno, level, btree_curs);
bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(agbno);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
XFS_FSB_TO_BB(mp, 1));
lptr->agbno = agbno;
if (lptr->modulo)
lptr->modulo--;
/*
* initialize block header
*/
lptr->buf_p->b_ops = &xfs_allocbt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, crc_magic, level,
0, agno, XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, magic, level,
0, agno, 0);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
/*
* propagate extent record for first extent in new block up
*/
prop_freespace_cursor(mp, agno, btree_curs, startblock,
blockcount, level, magic);
}
/*
* add extent info to current block
*/
be16_add_cpu(&bt_hdr->bb_numrecs, 1);
bt_key = XFS_ALLOC_KEY_ADDR(mp, bt_hdr,
be16_to_cpu(bt_hdr->bb_numrecs));
bt_ptr = XFS_ALLOC_PTR_ADDR(mp, bt_hdr,
be16_to_cpu(bt_hdr->bb_numrecs),
mp->m_alloc_mxr[1]);
bt_key->ar_startblock = cpu_to_be32(startblock);
bt_key->ar_blockcount = cpu_to_be32(blockcount);
*bt_ptr = cpu_to_be32(btree_curs->level[level-1].agbno);
}
/*
* 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;
}
/*
* 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;
}
xfs_daddr_t
xfs_agb_to_daddr(xfs_mount_t *mp, xfs_agnumber_t agno, xfs_agblock_t agbno)
{
return XFS_AGB_TO_DADDR(mp, agno, agbno);
}
/*
* 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;
}
/*
* Check that the holemask and freemask of a hypothetical inode cluster match
* what's actually on disk. If sparse inodes are enabled, the cluster does
* not actually have to map to inodes if the corresponding holemask bit is set.
*
* @cluster_base is the first inode in the cluster within the @irec.
*/
STATIC int
xchk_iallocbt_check_cluster(
struct xchk_btree *bs,
struct xfs_inobt_rec_incore *irec,
unsigned int cluster_base)
{
struct xfs_imap imap;
struct xfs_mount *mp = bs->cur->bc_mp;
struct xfs_dinode *dip;
struct xfs_buf *cluster_bp;
unsigned int nr_inodes;
xfs_agnumber_t agno = bs->cur->bc_private.a.agno;
xfs_agblock_t agbno;
unsigned int cluster_index;
uint16_t cluster_mask = 0;
uint16_t ir_holemask;
int error = 0;
nr_inodes = min_t(unsigned int, XFS_INODES_PER_CHUNK,
mp->m_inodes_per_cluster);
/* Map this inode cluster */
agbno = XFS_AGINO_TO_AGBNO(mp, irec->ir_startino + cluster_base);
/* Compute a bitmask for this cluster that can be used for holemask. */
for (cluster_index = 0;
cluster_index < nr_inodes;
cluster_index += XFS_INODES_PER_HOLEMASK_BIT)
cluster_mask |= XFS_INOBT_MASK((cluster_base + cluster_index) /
XFS_INODES_PER_HOLEMASK_BIT);
/*
* Map the first inode of this cluster to a buffer and offset.
* Be careful about inobt records that don't align with the start of
* the inode buffer when block sizes are large enough to hold multiple
* inode chunks. When this happens, cluster_base will be zero but
* ir_startino can be large enough to make im_boffset nonzero.
*/
ir_holemask = (irec->ir_holemask & cluster_mask);
imap.im_blkno = XFS_AGB_TO_DADDR(mp, agno, agbno);
imap.im_len = XFS_FSB_TO_BB(mp, mp->m_blocks_per_cluster);
imap.im_boffset = XFS_INO_TO_OFFSET(mp, irec->ir_startino);
if (imap.im_boffset != 0 && cluster_base != 0) {
ASSERT(imap.im_boffset == 0 || cluster_base == 0);
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
return 0;
}
trace_xchk_iallocbt_check_cluster(mp, agno, irec->ir_startino,
imap.im_blkno, imap.im_len, cluster_base, nr_inodes,
cluster_mask, ir_holemask,
XFS_INO_TO_OFFSET(mp, irec->ir_startino +
cluster_base));
/* The whole cluster must be a hole or not a hole. */
if (ir_holemask != cluster_mask && ir_holemask != 0) {
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
return 0;
}
/* If any part of this is a hole, skip it. */
if (ir_holemask) {
xchk_xref_is_not_owned_by(bs->sc, agbno,
mp->m_blocks_per_cluster,
&XFS_RMAP_OINFO_INODES);
return 0;
}
xchk_xref_is_owned_by(bs->sc, agbno, mp->m_blocks_per_cluster,
&XFS_RMAP_OINFO_INODES);
/* Grab the inode cluster buffer. */
error = xfs_imap_to_bp(mp, bs->cur->bc_tp, &imap, &dip, &cluster_bp,
0, 0);
if (!xchk_btree_xref_process_error(bs->sc, bs->cur, 0, &error))
return error;
/* Check free status of each inode within this cluster. */
for (cluster_index = 0; cluster_index < nr_inodes; cluster_index++) {
struct xfs_dinode *dip;
if (imap.im_boffset >= BBTOB(cluster_bp->b_length)) {
xchk_btree_set_corrupt(bs->sc, bs->cur, 0);
break;
}
dip = xfs_buf_offset(cluster_bp, imap.im_boffset);
error = xchk_iallocbt_check_cluster_ifree(bs, irec,
cluster_base + cluster_index, dip);
if (error)
break;
imap.im_boffset += mp->m_sb.sb_inodesize;
}
xfs_trans_brelse(bs->cur->bc_tp, cluster_bp);
return error;
}
static void
prop_ino_cursor(xfs_mount_t *mp, xfs_agnumber_t agno, bt_status_t *btree_curs,
xfs_agino_t startino, int level)
{
struct xfs_btree_block *bt_hdr;
xfs_inobt_key_t *bt_key;
xfs_inobt_ptr_t *bt_ptr;
xfs_agblock_t agbno;
bt_stat_level_t *lptr;
level++;
if (level >= btree_curs->num_levels)
return;
lptr = &btree_curs->level[level];
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
if (be16_to_cpu(bt_hdr->bb_numrecs) == 0) {
/*
* this only happens once to initialize the
* first path up the left side of the tree
* where the agbno's are already set up
*/
prop_ino_cursor(mp, agno, btree_curs, startino, level);
}
if (be16_to_cpu(bt_hdr->bb_numrecs) ==
lptr->num_recs_pb + (lptr->modulo > 0)) {
/*
* write out current prev block, grab us a new block,
* and set the rightsib pointer of current block
*/
#ifdef XR_BLD_INO_TRACE
fprintf(stderr, " ino prop agbno %d ", lptr->prev_agbno);
#endif
if (lptr->prev_agbno != NULLAGBLOCK) {
ASSERT(lptr->prev_buf_p != NULL);
libxfs_writebuf(lptr->prev_buf_p, 0);
}
lptr->prev_agbno = lptr->agbno;;
lptr->prev_buf_p = lptr->buf_p;
agbno = get_next_blockaddr(agno, level, btree_curs);
bt_hdr->bb_u.s.bb_rightsib = cpu_to_be32(agbno);
lptr->buf_p = libxfs_getbuf(mp->m_dev,
XFS_AGB_TO_DADDR(mp, agno, agbno),
XFS_FSB_TO_BB(mp, 1));
lptr->agbno = agbno;
if (lptr->modulo)
lptr->modulo--;
/*
* initialize block header
*/
lptr->buf_p->b_ops = &xfs_inobt_buf_ops;
bt_hdr = XFS_BUF_TO_BLOCK(lptr->buf_p);
memset(bt_hdr, 0, mp->m_sb.sb_blocksize);
if (xfs_sb_version_hascrc(&mp->m_sb))
xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_CRC_MAGIC,
level, 0, agno,
XFS_BTREE_CRC_BLOCKS);
else
xfs_btree_init_block(mp, lptr->buf_p, XFS_IBT_MAGIC,
level, 0, agno, 0);
bt_hdr->bb_u.s.bb_leftsib = cpu_to_be32(lptr->prev_agbno);
/*
* propagate extent record for first extent in new block up
*/
prop_ino_cursor(mp, agno, btree_curs, startino, level);
}
/*
* add inode info to current block
*/
be16_add_cpu(&bt_hdr->bb_numrecs, 1);
bt_key = XFS_INOBT_KEY_ADDR(mp, bt_hdr,
be16_to_cpu(bt_hdr->bb_numrecs));
bt_ptr = XFS_INOBT_PTR_ADDR(mp, bt_hdr,
be16_to_cpu(bt_hdr->bb_numrecs),
mp->m_inobt_mxr[1]);
bt_key->ir_startino = cpu_to_be32(startino);
*bt_ptr = cpu_to_be32(btree_curs->level[level-1].agbno);
}
