/* Dispose of every block of every extent in the bitmap. */
int
xrep_reap_extents(
struct xfs_scrub *sc,
struct xfs_bitmap *bitmap,
const struct xfs_owner_info *oinfo,
enum xfs_ag_resv_type type)
{
struct xfs_bitmap_range *bmr;
struct xfs_bitmap_range *n;
xfs_fsblock_t fsbno;
int error = 0;
ASSERT(xfs_sb_version_hasrmapbt(&sc->mp->m_sb));
for_each_xfs_bitmap_block(fsbno, bmr, n, bitmap) {
ASSERT(sc->ip != NULL ||
XFS_FSB_TO_AGNO(sc->mp, fsbno) == sc->sa.agno);
trace_xrep_dispose_btree_extent(sc->mp,
XFS_FSB_TO_AGNO(sc->mp, fsbno),
XFS_FSB_TO_AGBNO(sc->mp, fsbno), 1);
error = xrep_reap_block(sc, fsbno, oinfo, type);
if (error)
goto out;
}
/* Initialize a new AG btree root block with zero entries. */
int
xrep_init_btblock(
struct xfs_scrub *sc,
xfs_fsblock_t fsb,
struct xfs_buf **bpp,
xfs_btnum_t btnum,
const struct xfs_buf_ops *ops)
{
struct xfs_trans *tp = sc->tp;
struct xfs_mount *mp = sc->mp;
struct xfs_buf *bp;
trace_xrep_init_btblock(mp, XFS_FSB_TO_AGNO(mp, fsb),
XFS_FSB_TO_AGBNO(mp, fsb), btnum);
ASSERT(XFS_FSB_TO_AGNO(mp, fsb) == sc->sa.agno);
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, XFS_FSB_TO_DADDR(mp, fsb),
XFS_FSB_TO_BB(mp, 1), 0);
xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
xfs_btree_init_block(mp, bp, btnum, 0, 0, sc->sa.agno, 0);
xfs_trans_buf_set_type(tp, bp, XFS_BLFT_BTREE_BUF);
xfs_trans_log_buf(tp, bp, 0, bp->b_length);
bp->b_ops = ops;
*bpp = bp;
return 0;
}
int
xfs_ioc_trim(
struct xfs_mount *mp,
struct fstrim_range __user *urange)
{
struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
unsigned int granularity = q->limits.discard_granularity;
struct fstrim_range range;
xfs_fsblock_t start, len, minlen;
xfs_agnumber_t start_agno, end_agno, agno;
__uint64_t blocks_trimmed = 0;
int error, last_error = 0;
if (!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
if (!blk_queue_discard(q))
return -XFS_ERROR(EOPNOTSUPP);
if (copy_from_user(&range, urange, sizeof(range)))
return -XFS_ERROR(EFAULT);
/*
* Truncating down the len isn't actually quite correct, but using
* XFS_B_TO_FSB would mean we trivially get overflows for values
* of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default
* used by the fstrim application. In the end it really doesn't
* matter as trimming blocks is an advisory interface.
*/
start = XFS_B_TO_FSBT(mp, range.start);
len = XFS_B_TO_FSBT(mp, range.len);
minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen));
start_agno = XFS_FSB_TO_AGNO(mp, start);
if (start_agno >= mp->m_sb.sb_agcount)
return -XFS_ERROR(EINVAL);
end_agno = XFS_FSB_TO_AGNO(mp, start + len);
if (end_agno >= mp->m_sb.sb_agcount)
end_agno = mp->m_sb.sb_agcount - 1;
for (agno = start_agno; agno <= end_agno; agno++) {
error = -xfs_trim_extents(mp, agno, start, len, minlen,
&blocks_trimmed);
if (error)
last_error = error;
}
if (last_error)
return last_error;
range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
if (copy_to_user(urange, &range, sizeof(range)))
return -XFS_ERROR(EFAULT);
return 0;
}
/* Sort refcount intents by AG. */
static int
xfs_refcount_update_diff_items(
void *priv,
struct list_head *a,
struct list_head *b)
{
struct xfs_mount *mp = priv;
struct xfs_refcount_intent *ra;
struct xfs_refcount_intent *rb;
ra = container_of(a, struct xfs_refcount_intent, ri_list);
rb = container_of(b, struct xfs_refcount_intent, ri_list);
return XFS_FSB_TO_AGNO(mp, ra->ri_startblock) -
XFS_FSB_TO_AGNO(mp, rb->ri_startblock);
}
/*
* Record a rmap intent; the list is kept sorted first by AG and then by
* increasing age.
*/
static int
__xfs_rmap_add(
struct xfs_mount *mp,
struct xfs_defer_ops *dfops,
enum xfs_rmap_intent_type type,
__uint64_t owner,
int whichfork,
struct xfs_bmbt_irec *bmap)
{
struct xfs_rmap_intent *ri;
trace_xfs_rmap_defer(mp, XFS_FSB_TO_AGNO(mp, bmap->br_startblock),
type,
XFS_FSB_TO_AGBNO(mp, bmap->br_startblock),
owner, whichfork,
bmap->br_startoff,
bmap->br_blockcount,
bmap->br_state);
ri = kmem_alloc(sizeof(struct xfs_rmap_intent), KM_SLEEP | KM_NOFS);
INIT_LIST_HEAD(&ri->ri_list);
ri->ri_type = type;
ri->ri_owner = owner;
ri->ri_whichfork = whichfork;
ri->ri_bmap = *bmap;
xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_RMAP, &ri->ri_list);
return 0;
}
/*
* Trim the mapping to the next block where there's a change in the
* shared/unshared status. More specifically, this means that we
* find the lowest-numbered extent of shared blocks that coincides with
* the given block mapping. If the shared extent overlaps the start of
* the mapping, trim the mapping to the end of the shared extent. If
* the shared region intersects the mapping, trim the mapping to the
* start of the shared extent. If there are no shared regions that
* overlap, just return the original extent.
*/
int
xfs_reflink_trim_around_shared(
struct xfs_inode *ip,
struct xfs_bmbt_irec *irec,
bool *shared,
bool *trimmed)
{
xfs_agnumber_t agno;
xfs_agblock_t agbno;
xfs_extlen_t aglen;
xfs_agblock_t fbno;
xfs_extlen_t flen;
int error = 0;
/* Holes, unwritten, and delalloc extents cannot be shared */
if (!xfs_is_reflink_inode(ip) || !xfs_bmap_is_real_extent(irec)) {
*shared = false;
return 0;
}
trace_xfs_reflink_trim_around_shared(ip, irec);
agno = XFS_FSB_TO_AGNO(ip->i_mount, irec->br_startblock);
agbno = XFS_FSB_TO_AGBNO(ip->i_mount, irec->br_startblock);
aglen = irec->br_blockcount;
error = xfs_reflink_find_shared(ip->i_mount, NULL, agno, agbno,
aglen, &fbno, &flen, true);
if (error)
return error;
*shared = *trimmed = false;
if (fbno == NULLAGBLOCK) {
/* No shared blocks at all. */
return 0;
} else if (fbno == agbno) {
/*
* The start of this extent is shared. Truncate the
* mapping at the end of the shared region so that a
* subsequent iteration starts at the start of the
* unshared region.
*/
irec->br_blockcount = flen;
*shared = true;
if (flen != aglen)
*trimmed = true;
return 0;
} else {
/*
* There's a shared extent midway through this extent.
* Truncate the mapping at the start of the shared
* extent so that a subsequent iteration starts at the
* start of the shared region.
*/
irec->br_blockcount = fbno - agbno;
*trimmed = true;
return 0;
}
}
/*
* Verify that an FS block number pointer neither points outside the
* filesystem nor points at static AG metadata.
*/
bool
xfs_verify_fsbno(
struct xfs_mount *mp,
xfs_fsblock_t fsbno)
{
xfs_agnumber_t agno = XFS_FSB_TO_AGNO(mp, fsbno);
if (agno >= mp->m_sb.sb_agcount)
return false;
return xfs_verify_agbno(mp, agno, XFS_FSB_TO_AGBNO(mp, fsbno));
}
/* Does this inode need the reflink flag? */
int
xfs_reflink_inode_has_shared_extents(
struct xfs_trans *tp,
struct xfs_inode *ip,
bool *has_shared)
{
struct xfs_bmbt_irec got;
struct xfs_mount *mp = ip->i_mount;
struct xfs_ifork *ifp;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
xfs_extlen_t aglen;
xfs_agblock_t rbno;
xfs_extlen_t rlen;
xfs_extnum_t idx;
bool found;
int error;
ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
if (!(ifp->if_flags & XFS_IFEXTENTS)) {
error = xfs_iread_extents(tp, ip, XFS_DATA_FORK);
if (error)
return error;
}
*has_shared = false;
found = xfs_iext_lookup_extent(ip, ifp, 0, &idx, &got);
while (found) {
if (isnullstartblock(got.br_startblock) ||
got.br_state != XFS_EXT_NORM)
goto next;
agno = XFS_FSB_TO_AGNO(mp, got.br_startblock);
agbno = XFS_FSB_TO_AGBNO(mp, got.br_startblock);
aglen = got.br_blockcount;
error = xfs_reflink_find_shared(mp, tp, agno, agbno, aglen,
&rbno, &rlen, false);
if (error)
return error;
/* Is there still a shared block here? */
if (rbno != NULLAGBLOCK) {
*has_shared = true;
return 0;
}
next:
found = xfs_iext_get_extent(ifp, ++idx, &got);
}
return 0;
}
/*
* Count fsblocks of the given fork.
*/
int /* error */
xfs_bmap_count_blocks(
xfs_trans_t *tp, /* transaction pointer */
xfs_inode_t *ip, /* incore inode */
int whichfork, /* data or attr fork */
int *count) /* out: count of blocks */
{
struct xfs_btree_block *block; /* current btree block */
xfs_fsblock_t bno; /* block # of "block" */
xfs_ifork_t *ifp; /* fork structure */
int level; /* btree level, for checking */
xfs_mount_t *mp; /* file system mount structure */
__be64 *pp; /* pointer to block address */
bno = NULLFSBLOCK;
mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) {
xfs_bmap_count_leaves(ifp, 0,
ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t),
count);
return 0;
}
/*
* Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
*/
block = ifp->if_broot;
level = be16_to_cpu(block->bb_level);
ASSERT(level > 0);
pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
bno = be64_to_cpu(*pp);
ASSERT(bno != NULLDFSBNO);
ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
mp);
return XFS_ERROR(EFSCORRUPTED);
}
return 0;
}
void
reset_bmaps(xfs_mount_t *mp)
{
xfs_agnumber_t agno;
xfs_agblock_t ag_size;
int ag_hdr_block;
ag_hdr_block = howmany(4 * mp->m_sb.sb_sectsize, mp->m_sb.sb_blocksize);
ag_size = mp->m_sb.sb_agblocks;
for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
if (agno == mp->m_sb.sb_agcount - 1)
ag_size = (xfs_extlen_t)(mp->m_sb.sb_dblocks -
(xfs_drfsbno_t)mp->m_sb.sb_agblocks * agno);
#ifdef BTREE_STATS
if (btree_find(ag_bmap[agno], 0, NULL)) {
printf("ag_bmap[%d] btree stats:\n", i);
btree_print_stats(ag_bmap[agno], stdout);
}
#endif
/*
* We always insert an item for the first block having a
* given state. So the code below means:
*
* block 0..ag_hdr_block-1: XR_E_INUSE_FS
* ag_hdr_block..ag_size: XR_E_UNKNOWN
* ag_size... XR_E_BAD_STATE
*/
btree_clear(ag_bmap[agno]);
btree_insert(ag_bmap[agno], 0, &states[XR_E_INUSE_FS]);
btree_insert(ag_bmap[agno],
ag_hdr_block, &states[XR_E_UNKNOWN]);
btree_insert(ag_bmap[agno], ag_size, &states[XR_E_BAD_STATE]);
}
if (mp->m_sb.sb_logstart != 0) {
set_bmap_ext(XFS_FSB_TO_AGNO(mp, mp->m_sb.sb_logstart),
XFS_FSB_TO_AGBNO(mp, mp->m_sb.sb_logstart),
mp->m_sb.sb_logblocks, XR_E_INUSE_FS);
}
reset_rt_bmap();
}
static void
scan_lbtree(
xfs_fsblock_t root,
int nlevels,
scan_lbtree_f_t func,
extmap_t **extmapp,
typnm_t btype)
{
push_cur();
set_cur(&typtab[btype], XFS_FSB_TO_DADDR(mp, root), blkbb, DB_RING_IGN,
NULL);
if (iocur_top->data == NULL) {
dbprintf(_("can't read btree block %u/%u\n"),
XFS_FSB_TO_AGNO(mp, root),
XFS_FSB_TO_AGBNO(mp, root));
return;
}
(*func)(iocur_top->data, nlevels - 1, extmapp, btype);
pop_cur();
}
static int
bmap_f(
int argc,
char **argv)
{
int afork = 0;
bmap_ext_t be;
int c;
xfs_dfiloff_t co, cosave;
int dfork = 0;
xfs_dinode_t *dip;
xfs_dfiloff_t eo;
xfs_dfilblks_t len;
int nex;
char *p;
int whichfork;
if (iocur_top->ino == NULLFSINO) {
dbprintf(_("no current inode\n"));
return 0;
}
optind = 0;
if (argc) while ((c = getopt(argc, argv, "ad")) != EOF) {
switch (c) {
case 'a':
afork = 1;
break;
case 'd':
dfork = 1;
break;
default:
dbprintf(_("bad option for bmap command\n"));
return 0;
}
}
if (afork + dfork == 0) {
push_cur();
set_cur_inode(iocur_top->ino);
dip = iocur_top->data;
if (be32_to_cpu(dip->di_core.di_nextents))
dfork = 1;
if (be16_to_cpu(dip->di_core.di_anextents))
afork = 1;
pop_cur();
}
if (optind < argc) {
co = (xfs_dfiloff_t)strtoull(argv[optind], &p, 0);
if (*p != '\0') {
dbprintf(_("bad block number for bmap %s\n"),
argv[optind]);
return 0;
}
optind++;
if (optind < argc) {
len = (xfs_dfilblks_t)strtoull(argv[optind], &p, 0);
if (*p != '\0') {
dbprintf(_("bad len for bmap %s\n"), argv[optind]);
return 0;
}
eo = co + len - 1;
} else
eo = co;
} else {
co = 0;
eo = -1;
}
cosave = co;
for (whichfork = XFS_DATA_FORK;
whichfork <= XFS_ATTR_FORK;
whichfork++) {
if (whichfork == XFS_DATA_FORK && !dfork)
continue;
if (whichfork == XFS_ATTR_FORK && !afork)
continue;
for (;;) {
nex = 1;
bmap(co, eo - co + 1, whichfork, &nex, &be);
if (nex == 0)
break;
dbprintf(_("%s offset %lld startblock %llu (%u/%u) count "
"%llu flag %u\n"),
whichfork == XFS_DATA_FORK ? _("data") : _("attr"),
be.startoff, be.startblock,
XFS_FSB_TO_AGNO(mp, be.startblock),
XFS_FSB_TO_AGBNO(mp, be.startblock),
be.blockcount, be.flag);
co = be.startoff + be.blockcount;
}
co = cosave;
}
return 0;
}
xfs_agnumber_t
xfs_fsb_to_agno(xfs_mount_t *mp, xfs_fsblock_t fsbno)
{
return XFS_FSB_TO_AGNO(mp, fsbno);
}
/* Execute a getfsmap query against the regular data device. */
STATIC int
__xfs_getfsmap_datadev(
struct xfs_trans *tp,
struct xfs_fsmap *keys,
struct xfs_getfsmap_info *info,
int (*query_fn)(struct xfs_trans *,
struct xfs_getfsmap_info *,
struct xfs_btree_cur **,
void *),
void *priv)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_btree_cur *bt_cur = NULL;
xfs_fsblock_t start_fsb;
xfs_fsblock_t end_fsb;
xfs_agnumber_t start_ag;
xfs_agnumber_t end_ag;
xfs_daddr_t eofs;
int error = 0;
eofs = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks);
if (keys[0].fmr_physical >= eofs)
return 0;
if (keys[1].fmr_physical >= eofs)
keys[1].fmr_physical = eofs - 1;
start_fsb = XFS_DADDR_TO_FSB(mp, keys[0].fmr_physical);
end_fsb = XFS_DADDR_TO_FSB(mp, keys[1].fmr_physical);
/*
* Convert the fsmap low/high keys to AG based keys. Initialize
* low to the fsmap low key and max out the high key to the end
* of the AG.
*/
info->low.rm_startblock = XFS_FSB_TO_AGBNO(mp, start_fsb);
info->low.rm_offset = XFS_BB_TO_FSBT(mp, keys[0].fmr_offset);
error = xfs_fsmap_owner_to_rmap(&info->low, &keys[0]);
if (error)
return error;
info->low.rm_blockcount = 0;
xfs_getfsmap_set_irec_flags(&info->low, &keys[0]);
info->high.rm_startblock = -1U;
info->high.rm_owner = ULLONG_MAX;
info->high.rm_offset = ULLONG_MAX;
info->high.rm_blockcount = 0;
info->high.rm_flags = XFS_RMAP_KEY_FLAGS | XFS_RMAP_REC_FLAGS;
start_ag = XFS_FSB_TO_AGNO(mp, start_fsb);
end_ag = XFS_FSB_TO_AGNO(mp, end_fsb);
/* Query each AG */
for (info->agno = start_ag; info->agno <= end_ag; info->agno++) {
/*
* Set the AG high key from the fsmap high key if this
* is the last AG that we're querying.
*/
if (info->agno == end_ag) {
info->high.rm_startblock = XFS_FSB_TO_AGBNO(mp,
end_fsb);
info->high.rm_offset = XFS_BB_TO_FSBT(mp,
keys[1].fmr_offset);
error = xfs_fsmap_owner_to_rmap(&info->high, &keys[1]);
if (error)
goto err;
xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
}
if (bt_cur) {
xfs_btree_del_cursor(bt_cur, XFS_BTREE_NOERROR);
bt_cur = NULL;
xfs_trans_brelse(tp, info->agf_bp);
info->agf_bp = NULL;
}
error = xfs_alloc_read_agf(mp, tp, info->agno, 0,
&info->agf_bp);
if (error)
goto err;
trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
trace_xfs_fsmap_high_key(mp, info->dev, info->agno,
&info->high);
error = query_fn(tp, info, &bt_cur, priv);
if (error)
goto err;
/*
* Set the AG low key to the start of the AG prior to
* moving on to the next AG.
*/
if (info->agno == start_ag) {
info->low.rm_startblock = 0;
info->low.rm_owner = 0;
info->low.rm_offset = 0;
info->low.rm_flags = 0;
}
}
/* Report any gap at the end of the AG */
info->last = true;
error = query_fn(tp, info, &bt_cur, priv);
if (error)
goto err;
err:
if (bt_cur)
xfs_btree_del_cursor(bt_cur, error < 0 ? XFS_BTREE_ERROR :
XFS_BTREE_NOERROR);
if (info->agf_bp) {
xfs_trans_brelse(tp, info->agf_bp);
info->agf_bp = NULL;
}
return error;
}
/*
* Start processing for a leaf or fuller btree.
* A leaf directory is one where the attribute fork is too big for
* the inode but is small enough to fit into one btree block
* outside the inode. This code is modelled after process_leaf_dir_block.
*
* returns 0 if things are ok, 1 if bad (attributes needs to be junked)
* repair is set, if anything was changed, but attributes can live thru it
*/
static int
process_longform_attr(
xfs_mount_t *mp,
xfs_ino_t ino,
xfs_dinode_t *dip,
blkmap_t *blkmap,
int *repair) /* out - 1 if something was fixed */
{
xfs_attr_leafblock_t *leaf;
xfs_fsblock_t bno;
xfs_buf_t *bp;
xfs_dahash_t next_hashval;
int repairlinks = 0;
struct xfs_attr3_icleaf_hdr leafhdr;
int error;
*repair = 0;
bno = blkmap_get(blkmap, 0);
if ( bno == NULLFSBLOCK ) {
if (dip->di_aformat == XFS_DINODE_FMT_EXTENTS &&
be16_to_cpu(dip->di_anextents) == 0)
return(0); /* the kernel can handle this state */
do_warn(
_("block 0 of inode %" PRIu64 " attribute fork is missing\n"),
ino);
return(1);
}
/* FIX FOR bug 653709 -- EKN */
if (mp->m_sb.sb_agcount < XFS_FSB_TO_AGNO(mp, bno)) {
do_warn(
_("agno of attribute fork of inode %" PRIu64 " out of regular partition\n"), ino);
return(1);
}
bp = libxfs_readbuf(mp->m_dev, XFS_FSB_TO_DADDR(mp, bno),
XFS_FSB_TO_BB(mp, 1), 0, &xfs_da3_node_buf_ops);
if (!bp) {
do_warn(
_("can't read block 0 of inode %" PRIu64 " attribute fork\n"),
ino);
return(1);
}
if (bp->b_error == -EFSBADCRC)
(*repair)++;
/* is this block sane? */
if (__check_attr_header(mp, bp, ino)) {
*repair = 0;
libxfs_putbuf(bp);
return 1;
}
/* verify leaf block */
leaf = bp->b_addr;
xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &leafhdr, leaf);
/* check sibling pointers in leaf block or root block 0 before
* we have to release the btree block
*/
if (leafhdr.forw != 0 || leafhdr.back != 0) {
if (!no_modify) {
do_warn(
_("clearing forw/back pointers in block 0 for attributes in inode %" PRIu64 "\n"),
ino);
repairlinks = 1;
leafhdr.forw = 0;
leafhdr.back = 0;
xfs_attr3_leaf_hdr_to_disk(mp->m_attr_geo,
leaf, &leafhdr);
} else {
do_warn(
_("would clear forw/back pointers in block 0 for attributes in inode %" PRIu64 "\n"), ino);
}
}
/*
* use magic number to tell us what type of attribute this is.
* it's possible to have a node or leaf attribute in either an
* extent format or btree format attribute fork.
*/
switch (leafhdr.magic) {
case XFS_ATTR_LEAF_MAGIC: /* leaf-form attribute */
case XFS_ATTR3_LEAF_MAGIC:
if (process_leaf_attr_block(mp, leaf, 0, ino, blkmap,
0, &next_hashval, repair)) {
*repair = 0;
/* the block is bad. lose the attribute fork. */
libxfs_putbuf(bp);
return(1);
}
*repair = *repair || repairlinks;
break;
case XFS_DA_NODE_MAGIC: /* btree-form attribute */
case XFS_DA3_NODE_MAGIC:
/* must do this now, to release block 0 before the traversal */
if ((*repair || repairlinks) && !no_modify) {
*repair = 1;
libxfs_writebuf(bp, 0);
} else
libxfs_putbuf(bp);
error = process_node_attr(mp, ino, dip, blkmap); /* + repair */
if (error)
*repair = 0;
return error;
default:
do_warn(
_("bad attribute leaf magic # %#x for dir ino %" PRIu64 "\n"),
be16_to_cpu(leaf->hdr.info.magic), ino);
libxfs_putbuf(bp);
*repair = 0;
return(1);
}
if (*repair && !no_modify)
libxfs_writebuf(bp, 0);
else
libxfs_putbuf(bp);
return(0); /* repair may be set */
}
/* Dispose of a single block. */
STATIC int
xrep_reap_block(
struct xfs_scrub *sc,
xfs_fsblock_t fsbno,
const struct xfs_owner_info *oinfo,
enum xfs_ag_resv_type resv)
{
struct xfs_btree_cur *cur;
struct xfs_buf *agf_bp = NULL;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
bool has_other_rmap;
int error;
agno = XFS_FSB_TO_AGNO(sc->mp, fsbno);
agbno = XFS_FSB_TO_AGBNO(sc->mp, fsbno);
/*
* If we are repairing per-inode metadata, we need to read in the AGF
* buffer. Otherwise, we're repairing a per-AG structure, so reuse
* the AGF buffer that the setup functions already grabbed.
*/
if (sc->ip) {
error = xfs_alloc_read_agf(sc->mp, sc->tp, agno, 0, &agf_bp);
if (error)
return error;
if (!agf_bp)
return -ENOMEM;
} else {
agf_bp = sc->sa.agf_bp;
}
cur = xfs_rmapbt_init_cursor(sc->mp, sc->tp, agf_bp, agno);
/* Can we find any other rmappings? */
error = xfs_rmap_has_other_keys(cur, agbno, 1, oinfo, &has_other_rmap);
xfs_btree_del_cursor(cur, error);
if (error)
goto out_free;
/*
* If there are other rmappings, this block is cross linked and must
* not be freed. Remove the reverse mapping and move on. Otherwise,
* we were the only owner of the block, so free the extent, which will
* also remove the rmap.
*
* XXX: XFS doesn't support detecting the case where a single block
* metadata structure is crosslinked with a multi-block structure
* because the buffer cache doesn't detect aliasing problems, so we
* can't fix 100% of crosslinking problems (yet). The verifiers will
* blow on writeout, the filesystem will shut down, and the admin gets
* to run xfs_repair.
*/
if (has_other_rmap)
error = xfs_rmap_free(sc->tp, agf_bp, agno, agbno, 1, oinfo);
else if (resv == XFS_AG_RESV_AGFL)
error = xrep_put_freelist(sc, agbno);
else
error = xfs_free_extent(sc->tp, fsbno, 1, oinfo, resv);
if (agf_bp != sc->sa.agf_bp)
xfs_trans_brelse(sc->tp, agf_bp);
if (error)
return error;
if (sc->ip)
return xfs_trans_roll_inode(&sc->tp, sc->ip);
return xrep_roll_ag_trans(sc);
out_free:
if (agf_bp != sc->sa.agf_bp)
xfs_trans_brelse(sc->tp, agf_bp);
return error;
}
/*
* The user wants to preemptively CoW all shared blocks in this file,
* which enables us to turn off the reflink flag. Iterate all
* extents which are not prealloc/delalloc to see which ranges are
* mentioned in the refcount tree, then read those blocks into the
* pagecache, dirty them, fsync them back out, and then we can update
* the inode flag. What happens if we run out of memory? :)
*/
STATIC int
xfs_reflink_dirty_extents(
struct xfs_inode *ip,
xfs_fileoff_t fbno,
xfs_filblks_t end,
xfs_off_t isize)
{
struct xfs_mount *mp = ip->i_mount;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
xfs_extlen_t aglen;
xfs_agblock_t rbno;
xfs_extlen_t rlen;
xfs_off_t fpos;
xfs_off_t flen;
struct xfs_bmbt_irec map[2];
int nmaps;
int error = 0;
while (end - fbno > 0) {
nmaps = 1;
/*
* Look for extents in the file. Skip holes, delalloc, or
* unwritten extents; they can't be reflinked.
*/
error = xfs_bmapi_read(ip, fbno, end - fbno, map, &nmaps, 0);
if (error)
goto out;
if (nmaps == 0)
break;
if (!xfs_bmap_is_real_extent(&map[0]))
goto next;
map[1] = map[0];
while (map[1].br_blockcount) {
agno = XFS_FSB_TO_AGNO(mp, map[1].br_startblock);
agbno = XFS_FSB_TO_AGBNO(mp, map[1].br_startblock);
aglen = map[1].br_blockcount;
error = xfs_reflink_find_shared(mp, NULL, agno, agbno,
aglen, &rbno, &rlen, true);
if (error)
goto out;
if (rbno == NULLAGBLOCK)
break;
/* Dirty the pages */
xfs_iunlock(ip, XFS_ILOCK_EXCL);
fpos = XFS_FSB_TO_B(mp, map[1].br_startoff +
(rbno - agbno));
flen = XFS_FSB_TO_B(mp, rlen);
if (fpos + flen > isize)
flen = isize - fpos;
error = iomap_file_dirty(VFS_I(ip), fpos, flen,
&xfs_iomap_ops);
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (error)
goto out;
map[1].br_blockcount -= (rbno - agbno + rlen);
map[1].br_startoff += (rbno - agbno + rlen);
map[1].br_startblock += (rbno - agbno + rlen);
}
next:
fbno = map[0].br_startoff + map[0].br_blockcount;
}
out:
return error;
}
/*
* Unmap a range of blocks from a file, then map other blocks into the hole.
* The range to unmap is (destoff : destoff + srcioff + irec->br_blockcount).
* The extent irec is mapped into dest at irec->br_startoff.
*/
STATIC int
xfs_reflink_remap_extent(
struct xfs_inode *ip,
struct xfs_bmbt_irec *irec,
xfs_fileoff_t destoff,
xfs_off_t new_isize)
{
struct xfs_mount *mp = ip->i_mount;
bool real_extent = xfs_bmap_is_real_extent(irec);
struct xfs_trans *tp;
xfs_fsblock_t firstfsb;
unsigned int resblks;
struct xfs_defer_ops dfops;
struct xfs_bmbt_irec uirec;
xfs_filblks_t rlen;
xfs_filblks_t unmap_len;
xfs_off_t newlen;
int error;
unmap_len = irec->br_startoff + irec->br_blockcount - destoff;
trace_xfs_reflink_punch_range(ip, destoff, unmap_len);
/* No reflinking if we're low on space */
if (real_extent) {
error = xfs_reflink_ag_has_free_space(mp,
XFS_FSB_TO_AGNO(mp, irec->br_startblock));
if (error)
goto out;
}
/* Start a rolling transaction to switch the mappings */
resblks = XFS_EXTENTADD_SPACE_RES(ip->i_mount, XFS_DATA_FORK);
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp);
if (error)
goto out;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
/* If we're not just clearing space, then do we have enough quota? */
if (real_extent) {
error = xfs_trans_reserve_quota_nblks(tp, ip,
irec->br_blockcount, 0, XFS_QMOPT_RES_REGBLKS);
if (error)
goto out_cancel;
}
trace_xfs_reflink_remap(ip, irec->br_startoff,
irec->br_blockcount, irec->br_startblock);
/* Unmap the old blocks in the data fork. */
rlen = unmap_len;
while (rlen) {
xfs_defer_init(&dfops, &firstfsb);
error = __xfs_bunmapi(tp, ip, destoff, &rlen, 0, 1,
&firstfsb, &dfops);
if (error)
goto out_defer;
/*
* Trim the extent to whatever got unmapped.
* Remember, bunmapi works backwards.
*/
uirec.br_startblock = irec->br_startblock + rlen;
uirec.br_startoff = irec->br_startoff + rlen;
uirec.br_blockcount = unmap_len - rlen;
unmap_len = rlen;
/* If this isn't a real mapping, we're done. */
if (!real_extent || uirec.br_blockcount == 0)
goto next_extent;
trace_xfs_reflink_remap(ip, uirec.br_startoff,
uirec.br_blockcount, uirec.br_startblock);
/* Update the refcount tree */
error = xfs_refcount_increase_extent(mp, &dfops, &uirec);
if (error)
goto out_defer;
/* Map the new blocks into the data fork. */
error = xfs_bmap_map_extent(mp, &dfops, ip, &uirec);
if (error)
goto out_defer;
/* Update quota accounting. */
xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT,
uirec.br_blockcount);
/* Update dest isize if needed. */
newlen = XFS_FSB_TO_B(mp,
uirec.br_startoff + uirec.br_blockcount);
newlen = min_t(xfs_off_t, newlen, new_isize);
if (newlen > i_size_read(VFS_I(ip))) {
trace_xfs_reflink_update_inode_size(ip, newlen);
i_size_write(VFS_I(ip), newlen);
ip->i_d.di_size = newlen;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
}
next_extent:
/* Process all the deferred stuff. */
xfs_defer_ijoin(&dfops, ip);
error = xfs_defer_finish(&tp, &dfops);
if (error)
goto out_defer;
}
error = xfs_trans_commit(tp);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
goto out;
return 0;
out_defer:
xfs_defer_cancel(&dfops);
out_cancel:
xfs_trans_cancel(tp);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
out:
trace_xfs_reflink_remap_extent_error(ip, error, _RET_IP_);
return error;
}
/* Clear the inode reflink flag if there are no shared extents. */
int
xfs_reflink_clear_inode_flag(
struct xfs_inode *ip,
struct xfs_trans **tpp)
{
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t fbno;
xfs_filblks_t end;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
xfs_extlen_t aglen;
xfs_agblock_t rbno;
xfs_extlen_t rlen;
struct xfs_bmbt_irec map;
int nmaps;
int error = 0;
ASSERT(xfs_is_reflink_inode(ip));
fbno = 0;
end = XFS_B_TO_FSB(mp, i_size_read(VFS_I(ip)));
while (end - fbno > 0) {
nmaps = 1;
/*
* Look for extents in the file. Skip holes, delalloc, or
* unwritten extents; they can't be reflinked.
*/
error = xfs_bmapi_read(ip, fbno, end - fbno, &map, &nmaps, 0);
if (error)
return error;
if (nmaps == 0)
break;
if (!xfs_bmap_is_real_extent(&map))
goto next;
agno = XFS_FSB_TO_AGNO(mp, map.br_startblock);
agbno = XFS_FSB_TO_AGBNO(mp, map.br_startblock);
aglen = map.br_blockcount;
error = xfs_reflink_find_shared(mp, agno, agbno, aglen,
&rbno, &rlen, false);
if (error)
return error;
/* Is there still a shared block here? */
if (rbno != NULLAGBLOCK)
return 0;
next:
fbno = map.br_startoff + map.br_blockcount;
}
/*
* We didn't find any shared blocks so turn off the reflink flag.
* First, get rid of any leftover CoW mappings.
*/
error = xfs_reflink_cancel_cow_blocks(ip, tpp, 0, NULLFILEOFF, true);
if (error)
return error;
/* Clear the inode flag. */
trace_xfs_reflink_unset_inode_flag(ip);
ip->i_d.di_flags2 &= ~XFS_DIFLAG2_REFLINK;
xfs_inode_clear_cowblocks_tag(ip);
xfs_trans_ijoin(*tpp, ip, 0);
xfs_trans_log_inode(*tpp, ip, XFS_ILOG_CORE);
return error;
}
/*
* Process one of the deferred rmap operations. We pass back the
* btree cursor to maintain our lock on the rmapbt between calls.
* This saves time and eliminates a buffer deadlock between the
* superblock and the AGF because we'll always grab them in the same
* order.
*/
int
xfs_rmap_finish_one(
struct xfs_trans *tp,
enum xfs_rmap_intent_type type,
__uint64_t owner,
int whichfork,
xfs_fileoff_t startoff,
xfs_fsblock_t startblock,
xfs_filblks_t blockcount,
xfs_exntst_t state,
struct xfs_btree_cur **pcur)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_btree_cur *rcur;
struct xfs_buf *agbp = NULL;
int error = 0;
xfs_agnumber_t agno;
struct xfs_owner_info oinfo;
xfs_agblock_t bno;
bool unwritten;
agno = XFS_FSB_TO_AGNO(mp, startblock);
ASSERT(agno != NULLAGNUMBER);
bno = XFS_FSB_TO_AGBNO(mp, startblock);
trace_xfs_rmap_deferred(mp, agno, type, bno, owner, whichfork,
startoff, blockcount, state);
if (XFS_TEST_ERROR(false, mp,
XFS_ERRTAG_RMAP_FINISH_ONE,
XFS_RANDOM_RMAP_FINISH_ONE))
return -EIO;
/*
* If we haven't gotten a cursor or the cursor AG doesn't match
* the startblock, get one now.
*/
rcur = *pcur;
if (rcur != NULL && rcur->bc_private.a.agno != agno) {
xfs_rmap_finish_one_cleanup(tp, rcur, 0);
rcur = NULL;
*pcur = NULL;
}
if (rcur == NULL) {
/*
* Refresh the freelist before we start changing the
* rmapbt, because a shape change could cause us to
* allocate blocks.
*/
error = xfs_free_extent_fix_freelist(tp, agno, &agbp);
if (error)
return error;
if (!agbp)
return -EFSCORRUPTED;
rcur = xfs_rmapbt_init_cursor(mp, tp, agbp, agno);
if (!rcur) {
error = -ENOMEM;
goto out_cur;
}
}
*pcur = rcur;
xfs_rmap_ino_owner(&oinfo, owner, whichfork, startoff);
unwritten = state == XFS_EXT_UNWRITTEN;
bno = XFS_FSB_TO_AGBNO(rcur->bc_mp, startblock);
switch (type) {
case XFS_RMAP_ALLOC:
case XFS_RMAP_MAP:
error = xfs_rmap_map(rcur, bno, blockcount, unwritten, &oinfo);
break;
case XFS_RMAP_FREE:
case XFS_RMAP_UNMAP:
error = xfs_rmap_unmap(rcur, bno, blockcount, unwritten,
&oinfo);
break;
case XFS_RMAP_CONVERT:
error = xfs_rmap_convert(rcur, bno, blockcount, !unwritten,
&oinfo);
break;
default:
ASSERT(0);
error = -EFSCORRUPTED;
}
return error;
out_cur:
xfs_trans_brelse(tp, agbp);
return error;
}