From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-xfs@vger.kernel.org
Subject: Re: [PATCH 12/18] xfs: introduce the xfs_iext_cursor abstraction
Date: Tue, 31 Oct 2017 15:02:34 -0700 [thread overview]
Message-ID: <20171031220234.GT4911@magnolia> (raw)
In-Reply-To: <20171031142230.11755-13-hch@lst.de>
On Tue, Oct 31, 2017 at 04:22:24PM +0200, Christoph Hellwig wrote:
> Add a new xfs_iext_cursor structure to hide the direct extent map
> index manipulations. In addition to the existing lookup/get/insert/
> remove and update routines new primitives to get the first and last
> extent cursor, as well as moving up and down by one extent are
> provided. Also new are convenience to increment/decrement the
> cursor and retreive the new extent, as well as to peek into the
> previous/next extent without updating the cursor and last but not
> least a macro to iterate over all extents in a fork.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
> fs/xfs/libxfs/xfs_bmap.c | 432 ++++++++++++++++++++---------------------
> fs/xfs/libxfs/xfs_bmap.h | 12 +-
> fs/xfs/libxfs/xfs_inode_fork.c | 75 +++----
> fs/xfs/libxfs/xfs_inode_fork.h | 87 ++++++++-
> fs/xfs/libxfs/xfs_types.h | 3 +
> fs/xfs/scrub/bmap.c | 6 +-
> fs/xfs/scrub/dir.c | 14 +-
> fs/xfs/xfs_bmap_util.c | 12 +-
> fs/xfs/xfs_dir2_readdir.c | 8 +-
> fs/xfs/xfs_dquot.c | 4 +-
> fs/xfs/xfs_iomap.c | 13 +-
> fs/xfs/xfs_reflink.c | 56 +++---
> fs/xfs/xfs_trace.h | 12 +-
> 13 files changed, 401 insertions(+), 333 deletions(-)
>
> diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
> index 56482bf6280d..453dc1ae76ab 100644
> --- a/fs/xfs/libxfs/xfs_bmap.c
> +++ b/fs/xfs/libxfs/xfs_bmap.c
> @@ -671,8 +671,9 @@ xfs_bmap_extents_to_btree(
> xfs_bmbt_key_t *kp; /* root block key pointer */
> xfs_mount_t *mp; /* mount structure */
> xfs_bmbt_ptr_t *pp; /* root block address pointer */
> + struct xfs_iext_cursor ext;
> struct xfs_bmbt_irec rec;
> - xfs_extnum_t i = 0, cnt = 0;
> + xfs_extnum_t cnt = 0;
>
> mp = ip->i_mount;
> ASSERT(whichfork != XFS_COW_FORK);
> @@ -751,7 +752,7 @@ xfs_bmap_extents_to_btree(
> XFS_BTNUM_BMAP, 0, 0, ip->i_ino,
> XFS_BTREE_LONG_PTRS);
>
> - while (xfs_iext_get_extent(ifp, i++, &rec)) {
> + for_each_iext(ifp, &ext, &rec) {
I'm torn here, because everything else has "xfs" in the name somewhere.
Consistently done we end up with this uglyness:
xfs_iext_for_each_extent(ifp, &ext, &rec) {
But ... it's syntactic sugar applied to a for loop. I like sugar.
for_each_xfs_iext() { ... } ?
Not sure, I rarely add such things to header files. The English works
better in this second version, so I think I like this (for_each_iext).
Anyone have strong opinions about this?
> if (isnullstartblock(rec.br_startblock))
> continue;
> arp = XFS_BMBT_REC_ADDR(mp, ablock, 1 + cnt);
> @@ -827,6 +828,7 @@ xfs_bmap_local_to_extents(
> xfs_alloc_arg_t args; /* allocation arguments */
> xfs_buf_t *bp; /* buffer for extent block */
> struct xfs_bmbt_irec rec;
> + struct xfs_iext_cursor ext;
I was expecting an extent map cursor to have 'cur' in the name --
'ext' misleads me into thinking that 'ext' is an actual extent.
struct xfs_iext_cursor *icur;
Also, xfs_iext_insert names its cursor parameter *cur, so why ext here?
For now I'm going to assume that xfs_iext_{first,last,next,prev} will
actually do something with the ifp parameter in the near future.
--D
>
> /*
> * We don't want to deal with the case of keeping inode data inline yet.
> @@ -893,7 +895,8 @@ xfs_bmap_local_to_extents(
> rec.br_startblock = args.fsbno;
> rec.br_blockcount = 1;
> rec.br_state = XFS_EXT_NORM;
> - xfs_iext_insert(ip, 0, 1, &rec, 0);
> + xfs_iext_first(ifp, &ext);
> + xfs_iext_insert(ip, &ext, 1, &rec, 0);
>
> XFS_IFORK_NEXT_SET(ip, whichfork, 1);
> ip->i_d.di_nblocks = 1;
> @@ -1173,6 +1176,7 @@ xfs_iread_extents(
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> xfs_extnum_t nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
> struct xfs_btree_block *block = ifp->if_broot;
> + struct xfs_iext_cursor ext;
> xfs_fsblock_t bno;
> struct xfs_buf *bp;
> xfs_extnum_t i, j;
> @@ -1222,6 +1226,7 @@ xfs_iread_extents(
> * Here with bp and block set to the leftmost leaf node in the tree.
> */
> i = 0;
> + xfs_iext_first(ifp, &ext);
>
> /*
> * Loop over all leaf nodes. Copy information to the extent records.
> @@ -1263,7 +1268,8 @@ xfs_iread_extents(
> }
> trp->l0 = be64_to_cpu(frp->l0);
> trp->l1 = be64_to_cpu(frp->l1);
> - trace_xfs_read_extent(ip, i, state, _THIS_IP_);
> + trace_xfs_read_extent(ip, &ext, state, _THIS_IP_);
> + xfs_iext_next(ifp, &ext);
> }
> xfs_trans_brelse(tp, bp);
> bno = nextbno;
> @@ -1311,7 +1317,7 @@ xfs_bmap_first_unused(
> {
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx = 0;
> + struct xfs_iext_cursor ext;
> xfs_fileoff_t lastaddr = 0;
> xfs_fileoff_t lowest, max;
> int error;
> @@ -1332,7 +1338,7 @@ xfs_bmap_first_unused(
> }
>
> lowest = max = *first_unused;
> - while (xfs_iext_get_extent(ifp, idx++, &got)) {
> + for_each_iext(ifp, &ext, &got) {
> /*
> * See if the hole before this extent will work.
> */
> @@ -1362,7 +1368,7 @@ xfs_bmap_last_before(
> {
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> int error;
>
> switch (XFS_IFORK_FORMAT(ip, whichfork)) {
> @@ -1382,7 +1388,7 @@ xfs_bmap_last_before(
> return error;
> }
>
> - if (!xfs_iext_lookup_extent_before(ip, ifp, last_block, &idx, &got))
> + if (!xfs_iext_lookup_extent_before(ip, ifp, last_block, &ext, &got))
> *last_block = 0;
> return 0;
> }
> @@ -1396,8 +1402,8 @@ xfs_bmap_last_extent(
> int *is_empty)
> {
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> + struct xfs_iext_cursor ext;
> int error;
> - int nextents;
>
> if (!(ifp->if_flags & XFS_IFEXTENTS)) {
> error = xfs_iread_extents(tp, ip, whichfork);
> @@ -1405,14 +1411,11 @@ xfs_bmap_last_extent(
> return error;
> }
>
> - nextents = xfs_iext_count(ifp);
> - if (nextents == 0) {
> + xfs_iext_last(ifp, &ext);
> + if (!xfs_iext_get_extent(ifp, &ext, rec))
> *is_empty = 1;
> - return 0;
> - }
> -
> - xfs_iext_get_extent(ifp, nextents - 1, rec);
> - *is_empty = 0;
> + else
> + *is_empty = 0;
> return 0;
> }
>
> @@ -1500,6 +1503,7 @@ xfs_bmap_one_block(
> xfs_ifork_t *ifp; /* inode fork pointer */
> int rval; /* return value */
> xfs_bmbt_irec_t s; /* internal version of extent */
> + struct xfs_iext_cursor ext;
>
> #ifndef DEBUG
> if (whichfork == XFS_DATA_FORK)
> @@ -1511,7 +1515,8 @@ xfs_bmap_one_block(
> return 0;
> ifp = XFS_IFORK_PTR(ip, whichfork);
> ASSERT(ifp->if_flags & XFS_IFEXTENTS);
> - xfs_iext_get_extent(ifp, 0, &s);
> + xfs_iext_first(ifp, &ext);
> + xfs_iext_get_extent(ifp, &ext, &s);
> rval = s.br_startoff == 0 && s.br_blockcount == 1;
> if (rval && whichfork == XFS_DATA_FORK)
> ASSERT(XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize);
> @@ -1553,8 +1558,6 @@ xfs_bmap_add_extent_delay_real(
> nextents = (whichfork == XFS_COW_FORK ? &bma->ip->i_cnextents :
> &bma->ip->i_d.di_nextents);
>
> - ASSERT(bma->idx >= 0);
> - ASSERT(bma->idx <= xfs_iext_count(ifp));
> ASSERT(!isnullstartblock(new->br_startblock));
> ASSERT(!bma->cur ||
> (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
> @@ -1568,7 +1571,7 @@ xfs_bmap_add_extent_delay_real(
> /*
> * Set up a bunch of variables to make the tests simpler.
> */
> - xfs_iext_get_extent(ifp, bma->idx, &PREV);
> + xfs_iext_get_extent(ifp, &bma->ext, &PREV);
> new_endoff = new->br_startoff + new->br_blockcount;
> ASSERT(isnullstartblock(PREV.br_startblock));
> ASSERT(PREV.br_startoff <= new->br_startoff);
> @@ -1590,10 +1593,8 @@ xfs_bmap_add_extent_delay_real(
> * Check and set flags if this segment has a left neighbor.
> * Don't set contiguous if the combined extent would be too large.
> */
> - if (bma->idx > 0) {
> + if (xfs_iext_peek_prev_extent(ifp, &bma->ext, &LEFT)) {
> state |= BMAP_LEFT_VALID;
> - xfs_iext_get_extent(ifp, bma->idx - 1, &LEFT);
> -
> if (isnullstartblock(LEFT.br_startblock))
> state |= BMAP_LEFT_DELAY;
> }
> @@ -1610,10 +1611,8 @@ xfs_bmap_add_extent_delay_real(
> * Don't set contiguous if the combined extent would be too large.
> * Also check for all-three-contiguous being too large.
> */
> - if (bma->idx < xfs_iext_count(ifp) - 1) {
> + if (xfs_iext_peek_next_extent(ifp, &bma->ext, &RIGHT)) {
> state |= BMAP_RIGHT_VALID;
> - xfs_iext_get_extent(ifp, bma->idx + 1, &RIGHT);
> -
> if (isnullstartblock(RIGHT.br_startblock))
> state |= BMAP_RIGHT_DELAY;
> }
> @@ -1645,9 +1644,9 @@ xfs_bmap_add_extent_delay_real(
> */
> LEFT.br_blockcount += PREV.br_blockcount + RIGHT.br_blockcount;
>
> - xfs_iext_remove(bma->ip, bma->idx, 2, state);
> - bma->idx--;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &LEFT);
> + xfs_iext_remove(bma->ip, &bma->ext, 2, state);
> + xfs_iext_prev(ifp, &bma->ext);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &LEFT);
> (*nextents)--;
>
> if (bma->cur == NULL)
> @@ -1680,9 +1679,9 @@ xfs_bmap_add_extent_delay_real(
> old = LEFT;
> LEFT.br_blockcount += PREV.br_blockcount;
>
> - xfs_iext_remove(bma->ip, bma->idx, 1, state);
> - bma->idx--;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &LEFT);
> + xfs_iext_remove(bma->ip, &bma->ext, 1, state);
> + xfs_iext_prev(ifp, &bma->ext);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &LEFT);
>
> if (bma->cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -1706,10 +1705,10 @@ xfs_bmap_add_extent_delay_real(
> PREV.br_startblock = new->br_startblock;
> PREV.br_blockcount += RIGHT.br_blockcount;
>
> - bma->idx++;
> - xfs_iext_remove(bma->ip, bma->idx, 1, state);
> - bma->idx--;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &PREV);
> + xfs_iext_next(ifp, &bma->ext);
> + xfs_iext_remove(bma->ip, &bma->ext, 1, state);
> + xfs_iext_prev(ifp, &bma->ext);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &PREV);
>
> if (bma->cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -1733,7 +1732,7 @@ xfs_bmap_add_extent_delay_real(
> */
> PREV.br_startblock = new->br_startblock;
> PREV.br_state = new->br_state;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &PREV);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &PREV);
>
> (*nextents)++;
> if (bma->cur == NULL)
> @@ -1767,9 +1766,9 @@ xfs_bmap_add_extent_delay_real(
> PREV.br_startoff += new->br_blockcount;
> PREV.br_startblock = nullstartblock(da_new);
>
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &PREV);
> - bma->idx--;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &LEFT);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &PREV);
> + xfs_iext_prev(ifp, &bma->ext);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &LEFT);
>
> if (bma->cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -1783,7 +1782,6 @@ xfs_bmap_add_extent_delay_real(
> if (error)
> goto done;
> }
> -
> break;
>
> case BMAP_LEFT_FILLING:
> @@ -1791,7 +1789,7 @@ xfs_bmap_add_extent_delay_real(
> * Filling in the first part of a previous delayed allocation.
> * The left neighbor is not contiguous.
> */
> - xfs_iext_update_extent(bma->ip, state, bma->idx, new);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, new);
> (*nextents)++;
> if (bma->cur == NULL)
> rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
> @@ -1824,7 +1822,9 @@ xfs_bmap_add_extent_delay_real(
> PREV.br_startoff = new_endoff;
> PREV.br_blockcount = temp;
> PREV.br_startblock = nullstartblock(da_new);
> - xfs_iext_insert(bma->ip, bma->idx + 1, 1, &PREV, state);
> + xfs_iext_next(ifp, &bma->ext);
> + xfs_iext_insert(bma->ip, &bma->ext, 1, &PREV, state);
> + xfs_iext_prev(ifp, &bma->ext);
> break;
>
> case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
> @@ -1857,9 +1857,9 @@ xfs_bmap_add_extent_delay_real(
> PREV.br_blockcount = temp;
> PREV.br_startblock = nullstartblock(da_new);
>
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &PREV);
> - bma->idx++;
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &RIGHT);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &PREV);
> + xfs_iext_next(ifp, &bma->ext);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &RIGHT);
> break;
>
> case BMAP_RIGHT_FILLING:
> @@ -1867,7 +1867,7 @@ xfs_bmap_add_extent_delay_real(
> * Filling in the last part of a previous delayed allocation.
> * The right neighbor is not contiguous.
> */
> - xfs_iext_update_extent(bma->ip, state, bma->idx, new);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, new);
> (*nextents)++;
> if (bma->cur == NULL)
> rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
> @@ -1899,9 +1899,8 @@ xfs_bmap_add_extent_delay_real(
>
> PREV.br_startblock = nullstartblock(da_new);
> PREV.br_blockcount = temp;
> - xfs_iext_insert(bma->ip, bma->idx, 1, &PREV, state);
> -
> - bma->idx++;
> + xfs_iext_insert(bma->ip, &bma->ext, 1, &PREV, state);
> + xfs_iext_next(ifp, &bma->ext);
> break;
>
> case 0:
> @@ -1944,10 +1943,11 @@ xfs_bmap_add_extent_delay_real(
> PREV.br_startblock =
> nullstartblock(xfs_bmap_worst_indlen(bma->ip,
> PREV.br_blockcount));
> - xfs_iext_update_extent(bma->ip, state, bma->idx, &PREV);
> + xfs_iext_update_extent(bma->ip, state, &bma->ext, &PREV);
>
> /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */
> - xfs_iext_insert(bma->ip, bma->idx + 1, 2, &LEFT, state);
> + xfs_iext_next(ifp, &bma->ext);
> + xfs_iext_insert(bma->ip, &bma->ext, 2, &LEFT, state);
> (*nextents)++;
>
> if (bma->cur == NULL)
> @@ -1975,7 +1975,6 @@ xfs_bmap_add_extent_delay_real(
>
> da_new = startblockval(PREV.br_startblock) +
> startblockval(RIGHT.br_startblock);
> - bma->idx++;
> break;
>
> case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
> @@ -2039,7 +2038,7 @@ xfs_bmap_add_extent_unwritten_real(
> struct xfs_trans *tp,
> xfs_inode_t *ip, /* incore inode pointer */
> int whichfork,
> - xfs_extnum_t *idx, /* extent number to update/insert */
> + struct xfs_iext_cursor *ext,
> xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
> xfs_bmbt_irec_t *new, /* new data to add to file extents */
> xfs_fsblock_t *first, /* pointer to firstblock variable */
> @@ -2063,8 +2062,6 @@ xfs_bmap_add_extent_unwritten_real(
> cur = *curp;
> ifp = XFS_IFORK_PTR(ip, whichfork);
>
> - ASSERT(*idx >= 0);
> - ASSERT(*idx <= xfs_iext_count(ifp));
> ASSERT(!isnullstartblock(new->br_startblock));
>
> XFS_STATS_INC(mp, xs_add_exlist);
> @@ -2077,7 +2074,7 @@ xfs_bmap_add_extent_unwritten_real(
> * Set up a bunch of variables to make the tests simpler.
> */
> error = 0;
> - xfs_iext_get_extent(ifp, *idx, &PREV);
> + xfs_iext_get_extent(ifp, ext, &PREV);
> ASSERT(new->br_state != PREV.br_state);
> new_endoff = new->br_startoff + new->br_blockcount;
> ASSERT(PREV.br_startoff <= new->br_startoff);
> @@ -2096,10 +2093,8 @@ xfs_bmap_add_extent_unwritten_real(
> * Check and set flags if this segment has a left neighbor.
> * Don't set contiguous if the combined extent would be too large.
> */
> - if (*idx > 0) {
> + if (xfs_iext_peek_prev_extent(ifp, ext, &LEFT)) {
> state |= BMAP_LEFT_VALID;
> - xfs_iext_get_extent(ifp, *idx - 1, &LEFT);
> -
> if (isnullstartblock(LEFT.br_startblock))
> state |= BMAP_LEFT_DELAY;
> }
> @@ -2116,9 +2111,8 @@ xfs_bmap_add_extent_unwritten_real(
> * Don't set contiguous if the combined extent would be too large.
> * Also check for all-three-contiguous being too large.
> */
> - if (*idx < xfs_iext_count(ifp) - 1) {
> + if (xfs_iext_peek_next_extent(ifp, ext, &RIGHT)) {
> state |= BMAP_RIGHT_VALID;
> - xfs_iext_get_extent(ifp, *idx + 1, &RIGHT);
> if (isnullstartblock(RIGHT.br_startblock))
> state |= BMAP_RIGHT_DELAY;
> }
> @@ -2149,9 +2143,9 @@ xfs_bmap_add_extent_unwritten_real(
> */
> LEFT.br_blockcount += PREV.br_blockcount + RIGHT.br_blockcount;
>
> - xfs_iext_remove(ip, *idx, 2, state);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &LEFT);
> + xfs_iext_remove(ip, ext, 2, state);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &LEFT);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) - 2);
> if (cur == NULL)
> @@ -2187,9 +2181,9 @@ xfs_bmap_add_extent_unwritten_real(
> */
> LEFT.br_blockcount += PREV.br_blockcount;
>
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &LEFT);
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &LEFT);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
> if (cur == NULL)
> @@ -2220,10 +2214,10 @@ xfs_bmap_add_extent_unwritten_real(
> PREV.br_blockcount += RIGHT.br_blockcount;
> PREV.br_state = new->br_state;
>
> - ++*idx;
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
>
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
> @@ -2254,7 +2248,7 @@ xfs_bmap_add_extent_unwritten_real(
> * the new one.
> */
> PREV.br_state = new->br_state;
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
>
> if (cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -2282,9 +2276,9 @@ xfs_bmap_add_extent_unwritten_real(
> PREV.br_startblock += new->br_blockcount;
> PREV.br_blockcount -= new->br_blockcount;
>
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &LEFT);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &LEFT);
>
> if (cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -2316,8 +2310,8 @@ xfs_bmap_add_extent_unwritten_real(
> PREV.br_startblock += new->br_blockcount;
> PREV.br_blockcount -= new->br_blockcount;
>
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> - xfs_iext_insert(ip, *idx, 1, new, state);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
> + xfs_iext_insert(ip, ext, 1, new, state);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
> if (cur == NULL)
> @@ -2350,9 +2344,9 @@ xfs_bmap_add_extent_unwritten_real(
> RIGHT.br_startblock = new->br_startblock;
> RIGHT.br_blockcount += new->br_blockcount;
>
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> - ++*idx;
> - xfs_iext_update_extent(ip, state, *idx, &RIGHT);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &RIGHT);
>
> if (cur == NULL)
> rval = XFS_ILOG_DEXT;
> @@ -2382,9 +2376,9 @@ xfs_bmap_add_extent_unwritten_real(
> old = PREV;
> PREV.br_blockcount -= new->br_blockcount;
>
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> - ++*idx;
> - xfs_iext_insert(ip, *idx, 1, new, state);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_insert(ip, ext, 1, new, state);
>
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
> @@ -2425,9 +2419,9 @@ xfs_bmap_add_extent_unwritten_real(
> r[1].br_startblock = new->br_startblock + new->br_blockcount;
> r[1].br_state = PREV.br_state;
>
> - xfs_iext_update_extent(ip, state, *idx, &PREV);
> - ++*idx;
> - xfs_iext_insert(ip, *idx, 2, &r[0], state);
> + xfs_iext_update_extent(ip, state, ext, &PREV);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_insert(ip, ext, 2, &r[0], state);
>
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 2);
> @@ -2516,7 +2510,7 @@ STATIC void
> xfs_bmap_add_extent_hole_delay(
> xfs_inode_t *ip, /* incore inode pointer */
> int whichfork,
> - xfs_extnum_t *idx, /* extent number to update/insert */
> + struct xfs_iext_cursor *ext,
> xfs_bmbt_irec_t *new) /* new data to add to file extents */
> {
> xfs_ifork_t *ifp; /* inode fork pointer */
> @@ -2533,10 +2527,8 @@ xfs_bmap_add_extent_hole_delay(
> /*
> * Check and set flags if this segment has a left neighbor
> */
> - if (*idx > 0) {
> + if (xfs_iext_peek_prev_extent(ifp, ext, &left)) {
> state |= BMAP_LEFT_VALID;
> - xfs_iext_get_extent(ifp, *idx - 1, &left);
> -
> if (isnullstartblock(left.br_startblock))
> state |= BMAP_LEFT_DELAY;
> }
> @@ -2545,10 +2537,8 @@ xfs_bmap_add_extent_hole_delay(
> * Check and set flags if the current (right) segment exists.
> * If it doesn't exist, we're converting the hole at end-of-file.
> */
> - if (*idx < xfs_iext_count(ifp)) {
> + if (xfs_iext_get_extent(ifp, ext, &right)) {
> state |= BMAP_RIGHT_VALID;
> - xfs_iext_get_extent(ifp, *idx, &right);
> -
> if (isnullstartblock(right.br_startblock))
> state |= BMAP_RIGHT_DELAY;
> }
> @@ -2591,9 +2581,9 @@ xfs_bmap_add_extent_hole_delay(
> left.br_startblock = nullstartblock(newlen);
> left.br_blockcount = temp;
>
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &left);
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &left);
> break;
>
> case BMAP_LEFT_CONTIG:
> @@ -2611,8 +2601,8 @@ xfs_bmap_add_extent_hole_delay(
> left.br_blockcount = temp;
> left.br_startblock = nullstartblock(newlen);
>
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &left);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &left);
> break;
>
> case BMAP_RIGHT_CONTIG:
> @@ -2629,7 +2619,7 @@ xfs_bmap_add_extent_hole_delay(
> right.br_startoff = new->br_startoff;
> right.br_startblock = nullstartblock(newlen);
> right.br_blockcount = temp;
> - xfs_iext_update_extent(ip, state, *idx, &right);
> + xfs_iext_update_extent(ip, state, ext, &right);
> break;
>
> case 0:
> @@ -2639,7 +2629,7 @@ xfs_bmap_add_extent_hole_delay(
> * Insert a new entry.
> */
> oldlen = newlen = 0;
> - xfs_iext_insert(ip, *idx, 1, new, state);
> + xfs_iext_insert(ip, ext, 1, new, state);
> break;
> }
> if (oldlen != newlen) {
> @@ -2660,7 +2650,7 @@ xfs_bmap_add_extent_hole_real(
> struct xfs_trans *tp,
> struct xfs_inode *ip,
> int whichfork,
> - xfs_extnum_t *idx,
> + struct xfs_iext_cursor *ext,
> struct xfs_btree_cur **curp,
> struct xfs_bmbt_irec *new,
> xfs_fsblock_t *first,
> @@ -2678,8 +2668,6 @@ xfs_bmap_add_extent_hole_real(
> int state = xfs_bmap_fork_to_state(whichfork);
> struct xfs_bmbt_irec old;
>
> - ASSERT(*idx >= 0);
> - ASSERT(*idx <= xfs_iext_count(ifp));
> ASSERT(!isnullstartblock(new->br_startblock));
> ASSERT(!cur || !(cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
>
> @@ -2688,9 +2676,8 @@ xfs_bmap_add_extent_hole_real(
> /*
> * Check and set flags if this segment has a left neighbor.
> */
> - if (*idx > 0) {
> + if (xfs_iext_peek_prev_extent(ifp, ext, &left)) {
> state |= BMAP_LEFT_VALID;
> - xfs_iext_get_extent(ifp, *idx - 1, &left);
> if (isnullstartblock(left.br_startblock))
> state |= BMAP_LEFT_DELAY;
> }
> @@ -2699,9 +2686,8 @@ xfs_bmap_add_extent_hole_real(
> * Check and set flags if this segment has a current value.
> * Not true if we're inserting into the "hole" at eof.
> */
> - if (*idx < xfs_iext_count(ifp)) {
> + if (xfs_iext_get_extent(ifp, ext, &right)) {
> state |= BMAP_RIGHT_VALID;
> - xfs_iext_get_extent(ifp, *idx, &right);
> if (isnullstartblock(right.br_startblock))
> state |= BMAP_RIGHT_DELAY;
> }
> @@ -2740,9 +2726,9 @@ xfs_bmap_add_extent_hole_real(
> */
> left.br_blockcount += new->br_blockcount + right.br_blockcount;
>
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &left);
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &left);
>
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
> @@ -2777,8 +2763,8 @@ xfs_bmap_add_extent_hole_real(
> old = left;
> left.br_blockcount += new->br_blockcount;
>
> - --*idx;
> - xfs_iext_update_extent(ip, state, *idx, &left);
> + xfs_iext_prev(ifp, ext);
> + xfs_iext_update_extent(ip, state, ext, &left);
>
> if (cur == NULL) {
> rval = xfs_ilog_fext(whichfork);
> @@ -2805,7 +2791,7 @@ xfs_bmap_add_extent_hole_real(
> right.br_startoff = new->br_startoff;
> right.br_startblock = new->br_startblock;
> right.br_blockcount += new->br_blockcount;
> - xfs_iext_update_extent(ip, state, *idx, &right);
> + xfs_iext_update_extent(ip, state, ext, &right);
>
> if (cur == NULL) {
> rval = xfs_ilog_fext(whichfork);
> @@ -2827,7 +2813,7 @@ xfs_bmap_add_extent_hole_real(
> * real allocation.
> * Insert a new entry.
> */
> - xfs_iext_insert(ip, *idx, 1, new, state);
> + xfs_iext_insert(ip, ext, 1, new, state);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
> if (cur == NULL) {
> @@ -3777,7 +3763,7 @@ xfs_bmapi_read(
> struct xfs_bmbt_irec got;
> xfs_fileoff_t obno;
> xfs_fileoff_t end;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> int error;
> bool eof = false;
> int n = 0;
> @@ -3819,7 +3805,7 @@ xfs_bmapi_read(
> return error;
> }
>
> - if (!xfs_iext_lookup_extent(ip, ifp, bno, &idx, &got))
> + if (!xfs_iext_lookup_extent(ip, ifp, bno, &ext, &got))
> eof = true;
> end = bno + len;
> obno = bno;
> @@ -3851,7 +3837,7 @@ xfs_bmapi_read(
> break;
>
> /* Else go on to the next record. */
> - if (!xfs_iext_get_extent(ifp, ++idx, &got))
> + if (!xfs_iext_next_extent(ifp, &ext, &got))
> eof = true;
> }
> *nmap = n;
> @@ -3879,7 +3865,7 @@ xfs_bmapi_reserve_delalloc(
> xfs_filblks_t len,
> xfs_filblks_t prealloc,
> struct xfs_bmbt_irec *got,
> - xfs_extnum_t *lastx,
> + struct xfs_iext_cursor *ext,
> int eof)
> {
> struct xfs_mount *mp = ip->i_mount;
> @@ -3909,7 +3895,7 @@ xfs_bmapi_reserve_delalloc(
> if (extsz) {
> struct xfs_bmbt_irec prev;
>
> - if (!xfs_iext_get_extent(ifp, *lastx - 1, &prev))
> + if (!xfs_iext_peek_prev_extent(ifp, ext, &prev))
> prev.br_startoff = NULLFILEOFF;
>
> error = xfs_bmap_extsize_align(mp, got, &prev, extsz, rt, eof,
> @@ -3958,7 +3944,7 @@ xfs_bmapi_reserve_delalloc(
> got->br_blockcount = alen;
> got->br_state = XFS_EXT_NORM;
>
> - xfs_bmap_add_extent_hole_delay(ip, whichfork, lastx, got);
> + xfs_bmap_add_extent_hole_delay(ip, whichfork, ext, got);
>
> /*
> * Tag the inode if blocks were preallocated. Note that COW fork
> @@ -4003,8 +3989,7 @@ xfs_bmapi_allocate(
> if (bma->wasdel) {
> bma->length = (xfs_extlen_t)bma->got.br_blockcount;
> bma->offset = bma->got.br_startoff;
> - if (bma->idx)
> - xfs_iext_get_extent(ifp, bma->idx - 1, &bma->prev);
> + xfs_iext_peek_prev_extent(ifp, &bma->ext, &bma->prev);
> } else {
> bma->length = XFS_FILBLKS_MIN(bma->length, MAXEXTLEN);
> if (!bma->eof)
> @@ -4089,7 +4074,7 @@ xfs_bmapi_allocate(
> error = xfs_bmap_add_extent_delay_real(bma, whichfork);
> else
> error = xfs_bmap_add_extent_hole_real(bma->tp, bma->ip,
> - whichfork, &bma->idx, &bma->cur, &bma->got,
> + whichfork, &bma->ext, &bma->cur, &bma->got,
> bma->firstblock, bma->dfops, &bma->logflags);
>
> bma->logflags |= tmp_logflags;
> @@ -4101,7 +4086,7 @@ xfs_bmapi_allocate(
> * or xfs_bmap_add_extent_hole_real might have merged it into one of
> * the neighbouring ones.
> */
> - xfs_iext_get_extent(ifp, bma->idx, &bma->got);
> + xfs_iext_get_extent(ifp, &bma->ext, &bma->got);
>
> ASSERT(bma->got.br_startoff <= bma->offset);
> ASSERT(bma->got.br_startoff + bma->got.br_blockcount >=
> @@ -4159,7 +4144,7 @@ xfs_bmapi_convert_unwritten(
> }
>
> error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, whichfork,
> - &bma->idx, &bma->cur, mval, bma->firstblock, bma->dfops,
> + &bma->ext, &bma->cur, mval, bma->firstblock, bma->dfops,
> &tmp_logflags);
> /*
> * Log the inode core unconditionally in the unwritten extent conversion
> @@ -4182,7 +4167,7 @@ xfs_bmapi_convert_unwritten(
> * xfs_bmap_add_extent_unwritten_real might have merged it into one
> * of the neighbouring ones.
> */
> - xfs_iext_get_extent(ifp, bma->idx, &bma->got);
> + xfs_iext_get_extent(ifp, &bma->ext, &bma->got);
>
> /*
> * We may have combined previously unwritten space with written space,
> @@ -4301,9 +4286,9 @@ xfs_bmapi_write(
> end = bno + len;
> obno = bno;
>
> - if (!xfs_iext_lookup_extent(ip, ifp, bno, &bma.idx, &bma.got))
> + if (!xfs_iext_lookup_extent(ip, ifp, bno, &bma.ext, &bma.got))
> eof = true;
> - if (!xfs_iext_get_extent(ifp, bma.idx - 1, &bma.prev))
> + if (!xfs_iext_peek_prev_extent(ifp, &bma.ext, &bma.prev))
> bma.prev.br_startoff = NULLFILEOFF;
> bma.tp = tp;
> bma.ip = ip;
> @@ -4408,7 +4393,7 @@ xfs_bmapi_write(
>
> /* Else go on to the next record. */
> bma.prev = bma.got;
> - if (!xfs_iext_get_extent(ifp, ++bma.idx, &bma.got))
> + if (!xfs_iext_next_extent(ifp, &bma.ext, &bma.got))
> eof = true;
> }
> *nmap = n;
> @@ -4481,7 +4466,7 @@ xfs_bmapi_remap(
> struct xfs_btree_cur *cur = NULL;
> xfs_fsblock_t firstblock = NULLFSBLOCK;
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> int logflags = 0, error;
>
> ASSERT(len > 0);
> @@ -4505,7 +4490,7 @@ xfs_bmapi_remap(
> return error;
> }
>
> - if (xfs_iext_lookup_extent(ip, ifp, bno, &idx, &got)) {
> + if (xfs_iext_lookup_extent(ip, ifp, bno, &ext, &got)) {
> /* make sure we only reflink into a hole. */
> ASSERT(got.br_startoff > bno);
> ASSERT(got.br_startoff - bno >= len);
> @@ -4526,7 +4511,7 @@ xfs_bmapi_remap(
> got.br_blockcount = len;
> got.br_state = XFS_EXT_NORM;
>
> - error = xfs_bmap_add_extent_hole_real(tp, ip, XFS_DATA_FORK, &idx, &cur,
> + error = xfs_bmap_add_extent_hole_real(tp, ip, XFS_DATA_FORK, &ext, &cur,
> &got, &firstblock, dfops, &logflags);
> if (error)
> goto error0;
> @@ -4643,7 +4628,7 @@ int
> xfs_bmap_del_extent_delay(
> struct xfs_inode *ip,
> int whichfork,
> - xfs_extnum_t *idx,
> + struct xfs_iext_cursor *ext,
> struct xfs_bmbt_irec *got,
> struct xfs_bmbt_irec *del)
> {
> @@ -4665,8 +4650,6 @@ xfs_bmap_del_extent_delay(
> da_old = startblockval(got->br_startblock);
> da_new = 0;
>
> - ASSERT(*idx >= 0);
> - ASSERT(*idx <= xfs_iext_count(ifp));
> ASSERT(del->br_blockcount > 0);
> ASSERT(got->br_startoff <= del->br_startoff);
> ASSERT(got_endoff >= del_endoff);
> @@ -4700,8 +4683,8 @@ xfs_bmap_del_extent_delay(
> /*
> * Matches the whole extent. Delete the entry.
> */
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> break;
> case BMAP_LEFT_FILLING:
> /*
> @@ -4712,7 +4695,7 @@ xfs_bmap_del_extent_delay(
> da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip,
> got->br_blockcount), da_old);
> got->br_startblock = nullstartblock((int)da_new);
> - xfs_iext_update_extent(ip, state, *idx, got);
> + xfs_iext_update_extent(ip, state, ext, got);
> break;
> case BMAP_RIGHT_FILLING:
> /*
> @@ -4722,7 +4705,7 @@ xfs_bmap_del_extent_delay(
> da_new = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip,
> got->br_blockcount), da_old);
> got->br_startblock = nullstartblock((int)da_new);
> - xfs_iext_update_extent(ip, state, *idx, got);
> + xfs_iext_update_extent(ip, state, ext, got);
> break;
> case 0:
> /*
> @@ -4750,9 +4733,9 @@ xfs_bmap_del_extent_delay(
> new.br_state = got->br_state;
> new.br_startblock = nullstartblock((int)new_indlen);
>
> - xfs_iext_update_extent(ip, state, *idx, got);
> - ++*idx;
> - xfs_iext_insert(ip, *idx, 1, &new, state);
> + xfs_iext_update_extent(ip, state, ext, got);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_insert(ip, ext, 1, &new, state);
>
> da_new = got_indlen + new_indlen - stolen;
> del->br_blockcount -= stolen;
> @@ -4771,7 +4754,7 @@ xfs_bmap_del_extent_delay(
> void
> xfs_bmap_del_extent_cow(
> struct xfs_inode *ip,
> - xfs_extnum_t *idx,
> + struct xfs_iext_cursor *ext,
> struct xfs_bmbt_irec *got,
> struct xfs_bmbt_irec *del)
> {
> @@ -4786,8 +4769,6 @@ xfs_bmap_del_extent_cow(
> del_endoff = del->br_startoff + del->br_blockcount;
> got_endoff = got->br_startoff + got->br_blockcount;
>
> - ASSERT(*idx >= 0);
> - ASSERT(*idx <= xfs_iext_count(ifp));
> ASSERT(del->br_blockcount > 0);
> ASSERT(got->br_startoff <= del->br_startoff);
> ASSERT(got_endoff >= del_endoff);
> @@ -4803,8 +4784,8 @@ xfs_bmap_del_extent_cow(
> /*
> * Matches the whole extent. Delete the entry.
> */
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> break;
> case BMAP_LEFT_FILLING:
> /*
> @@ -4813,14 +4794,14 @@ xfs_bmap_del_extent_cow(
> got->br_startoff = del_endoff;
> got->br_blockcount -= del->br_blockcount;
> got->br_startblock = del->br_startblock + del->br_blockcount;
> - xfs_iext_update_extent(ip, state, *idx, got);
> + xfs_iext_update_extent(ip, state, ext, got);
> break;
> case BMAP_RIGHT_FILLING:
> /*
> * Deleting the last part of the extent.
> */
> got->br_blockcount -= del->br_blockcount;
> - xfs_iext_update_extent(ip, state, *idx, got);
> + xfs_iext_update_extent(ip, state, ext, got);
> break;
> case 0:
> /*
> @@ -4833,9 +4814,9 @@ xfs_bmap_del_extent_cow(
> new.br_state = got->br_state;
> new.br_startblock = del->br_startblock + del->br_blockcount;
>
> - xfs_iext_update_extent(ip, state, *idx, got);
> - ++*idx;
> - xfs_iext_insert(ip, *idx, 1, &new, state);
> + xfs_iext_update_extent(ip, state, ext, got);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_insert(ip, ext, 1, &new, state);
> break;
> }
> }
> @@ -4848,7 +4829,7 @@ STATIC int /* error */
> xfs_bmap_del_extent_real(
> xfs_inode_t *ip, /* incore inode pointer */
> xfs_trans_t *tp, /* current transaction pointer */
> - xfs_extnum_t *idx, /* extent number to update/delete */
> + struct xfs_iext_cursor *ext,
> struct xfs_defer_ops *dfops, /* list of extents to be freed */
> xfs_btree_cur_t *cur, /* if null, not a btree */
> xfs_bmbt_irec_t *del, /* data to remove from extents */
> @@ -4877,9 +4858,8 @@ xfs_bmap_del_extent_real(
> XFS_STATS_INC(mp, xs_del_exlist);
>
> ifp = XFS_IFORK_PTR(ip, whichfork);
> - ASSERT((*idx >= 0) && (*idx < xfs_iext_count(ifp)));
> ASSERT(del->br_blockcount > 0);
> - xfs_iext_get_extent(ifp, *idx, &got);
> + xfs_iext_get_extent(ifp, ext, &got);
> ASSERT(got.br_startoff <= del->br_startoff);
> del_endoff = del->br_startoff + del->br_blockcount;
> got_endoff = got.br_startoff + got.br_blockcount;
> @@ -4944,9 +4924,8 @@ xfs_bmap_del_extent_real(
> /*
> * Matches the whole extent. Delete the entry.
> */
> - xfs_iext_remove(ip, *idx, 1, state);
> - --*idx;
> -
> + xfs_iext_remove(ip, ext, 1, state);
> + xfs_iext_prev(ifp, ext);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
> flags |= XFS_ILOG_CORE;
> @@ -4965,7 +4944,7 @@ xfs_bmap_del_extent_real(
> got.br_startoff = del_endoff;
> got.br_startblock = del_endblock;
> got.br_blockcount -= del->br_blockcount;
> - xfs_iext_update_extent(ip, state, *idx, &got);
> + xfs_iext_update_extent(ip, state, ext, &got);
> if (!cur) {
> flags |= xfs_ilog_fext(whichfork);
> break;
> @@ -4979,7 +4958,7 @@ xfs_bmap_del_extent_real(
> * Deleting the last part of the extent.
> */
> got.br_blockcount -= del->br_blockcount;
> - xfs_iext_update_extent(ip, state, *idx, &got);
> + xfs_iext_update_extent(ip, state, ext, &got);
> if (!cur) {
> flags |= xfs_ilog_fext(whichfork);
> break;
> @@ -4995,7 +4974,7 @@ xfs_bmap_del_extent_real(
> old = got;
>
> got.br_blockcount = del->br_startoff - got.br_startoff;
> - xfs_iext_update_extent(ip, state, *idx, &got);
> + xfs_iext_update_extent(ip, state, ext, &got);
>
> new.br_startoff = del_endoff;
> new.br_blockcount = got_endoff - del_endoff;
> @@ -5039,7 +5018,7 @@ xfs_bmap_del_extent_real(
> * Reset the extent record back
> * to the original value.
> */
> - xfs_iext_update_extent(ip, state, *idx, &old);
> + xfs_iext_update_extent(ip, state, ext, &old);
> flags = 0;
> error = -ENOSPC;
> goto done;
> @@ -5049,8 +5028,8 @@ xfs_bmap_del_extent_real(
> flags |= xfs_ilog_fext(whichfork);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
> - ++*idx;
> - xfs_iext_insert(ip, *idx, 1, &new, state);
> + xfs_iext_next(ifp, ext);
> + xfs_iext_insert(ip, ext, 1, &new, state);
> break;
> }
>
> @@ -5113,7 +5092,6 @@ __xfs_bunmapi(
> xfs_bmbt_irec_t got; /* current extent record */
> xfs_ifork_t *ifp; /* inode fork pointer */
> int isrt; /* freeing in rt area */
> - xfs_extnum_t lastx; /* last extent index used */
> int logflags; /* transaction logging flags */
> xfs_extlen_t mod; /* rt extent offset */
> xfs_mount_t *mp; /* mount structure */
> @@ -5125,6 +5103,8 @@ __xfs_bunmapi(
> xfs_fileoff_t max_len;
> xfs_agnumber_t prev_agno = NULLAGNUMBER, agno;
> xfs_fileoff_t end;
> + struct xfs_iext_cursor ext;
> + bool done = false;
>
> trace_xfs_bunmap(ip, start, len, flags, _RET_IP_);
>
> @@ -5167,7 +5147,7 @@ __xfs_bunmapi(
> isrt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
> end = start + len;
>
> - if (!xfs_iext_lookup_extent_before(ip, ifp, &end, &lastx, &got)) {
> + if (!xfs_iext_lookup_extent_before(ip, ifp, &end, &ext, &got)) {
> *rlen = 0;
> return 0;
> }
> @@ -5194,16 +5174,16 @@ __xfs_bunmapi(
> }
>
> extno = 0;
> - while (end != (xfs_fileoff_t)-1 && end >= start && lastx >= 0 &&
> + while (end != (xfs_fileoff_t)-1 && end >= start &&
> (nexts == 0 || extno < nexts) && max_len > 0) {
> /*
> * Is the found extent after a hole in which end lives?
> * Just back up to the previous extent, if so.
> */
> - if (got.br_startoff > end) {
> - if (--lastx < 0)
> - break;
> - xfs_iext_get_extent(ifp, lastx, &got);
> + if (got.br_startoff > end &&
> + !xfs_iext_prev_extent(ifp, &ext, &got)) {
> + done = true;
> + break;
> }
> /*
> * Is the last block of this extent before the range
> @@ -5266,10 +5246,10 @@ __xfs_bunmapi(
> ASSERT(end >= mod);
> end -= mod > del.br_blockcount ?
> del.br_blockcount : mod;
> - if (end < got.br_startoff) {
> - if (--lastx >= 0)
> - xfs_iext_get_extent(ifp, lastx,
> - &got);
> + if (end < got.br_startoff &&
> + !xfs_iext_prev_extent(ifp, &ext, &got)) {
> + done = true;
> + break;
> }
> continue;
> }
> @@ -5290,7 +5270,7 @@ __xfs_bunmapi(
> }
> del.br_state = XFS_EXT_UNWRITTEN;
> error = xfs_bmap_add_extent_unwritten_real(tp, ip,
> - whichfork, &lastx, &cur, &del,
> + whichfork, &ext, &cur, &del,
> firstblock, dfops, &logflags);
> if (error)
> goto error0;
> @@ -5317,8 +5297,11 @@ __xfs_bunmapi(
> */
> ASSERT(end >= del.br_blockcount);
> end -= del.br_blockcount;
> - if (got.br_startoff > end && --lastx >= 0)
> - xfs_iext_get_extent(ifp, lastx, &got);
> + if (got.br_startoff > end &&
> + !xfs_iext_prev_extent(ifp, &ext, &got)) {
> + done = true;
> + break;
> + }
> continue;
> } else if (del.br_state == XFS_EXT_UNWRITTEN) {
> struct xfs_bmbt_irec prev;
> @@ -5329,8 +5312,8 @@ __xfs_bunmapi(
> * Unwrite the killed part of that one and
> * try again.
> */
> - ASSERT(lastx > 0);
> - xfs_iext_get_extent(ifp, lastx - 1, &prev);
> + if (!xfs_iext_prev_extent(ifp, &ext, &prev))
> + ASSERT(0);
> ASSERT(prev.br_state == XFS_EXT_NORM);
> ASSERT(!isnullstartblock(prev.br_startblock));
> ASSERT(del.br_startblock ==
> @@ -5342,9 +5325,8 @@ __xfs_bunmapi(
> prev.br_startoff = start;
> }
> prev.br_state = XFS_EXT_UNWRITTEN;
> - lastx--;
> error = xfs_bmap_add_extent_unwritten_real(tp,
> - ip, whichfork, &lastx, &cur,
> + ip, whichfork, &ext, &cur,
> &prev, firstblock, dfops,
> &logflags);
> if (error)
> @@ -5354,7 +5336,7 @@ __xfs_bunmapi(
> ASSERT(del.br_state == XFS_EXT_NORM);
> del.br_state = XFS_EXT_UNWRITTEN;
> error = xfs_bmap_add_extent_unwritten_real(tp,
> - ip, whichfork, &lastx, &cur,
> + ip, whichfork, &ext, &cur,
> &del, firstblock, dfops,
> &logflags);
> if (error)
> @@ -5364,10 +5346,10 @@ __xfs_bunmapi(
> }
>
> if (wasdel) {
> - error = xfs_bmap_del_extent_delay(ip, whichfork, &lastx,
> + error = xfs_bmap_del_extent_delay(ip, whichfork, &ext,
> &got, &del);
> } else {
> - error = xfs_bmap_del_extent_real(ip, tp, &lastx, dfops,
> + error = xfs_bmap_del_extent_real(ip, tp, &ext, dfops,
> cur, &del, &tmp_logflags, whichfork,
> flags);
> logflags |= tmp_logflags;
> @@ -5383,15 +5365,16 @@ __xfs_bunmapi(
> * If not done go on to the next (previous) record.
> */
> if (end != (xfs_fileoff_t)-1 && end >= start) {
> - if (lastx >= 0) {
> - xfs_iext_get_extent(ifp, lastx, &got);
> - if (got.br_startoff > end && --lastx >= 0)
> - xfs_iext_get_extent(ifp, lastx, &got);
> + if (!xfs_iext_get_extent(ifp, &ext, &got) ||
> + (got.br_startoff > end &&
> + !xfs_iext_prev_extent(ifp, &ext, &got))) {
> + done = true;
> + break;
> }
> extno++;
> }
> }
> - if (end == (xfs_fileoff_t)-1 || end < start || lastx < 0)
> + if (done || end == (xfs_fileoff_t)-1 || end < start)
> *rlen = 0;
> else
> *rlen = end - start + 1;
> @@ -5512,7 +5495,7 @@ xfs_bmse_merge(
> struct xfs_inode *ip,
> int whichfork,
> xfs_fileoff_t shift, /* shift fsb */
> - int *current_ext, /* idx of gotp */
> + struct xfs_iext_cursor *ext,
> struct xfs_bmbt_irec *got, /* extent to shift */
> struct xfs_bmbt_irec *left, /* preceding extent */
> struct xfs_btree_cur *cur,
> @@ -5567,10 +5550,10 @@ xfs_bmse_merge(
> return error;
>
> done:
> - xfs_iext_remove(ip, *current_ext, 1, 0);
> - --*current_ext;
> - xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork),
> - *current_ext, &new);
> + xfs_iext_remove(ip, ext, 1, 0);
> + xfs_iext_prev(XFS_IFORK_PTR(ip, whichfork), ext);
> + xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork), ext,
> + &new);
>
> /* update reverse mapping. rmap functions merge the rmaps for us */
> error = xfs_rmap_unmap_extent(mp, dfops, ip, whichfork, got);
> @@ -5585,7 +5568,7 @@ static int
> xfs_bmap_shift_update_extent(
> struct xfs_inode *ip,
> int whichfork,
> - xfs_extnum_t idx,
> + struct xfs_iext_cursor *ext,
> struct xfs_bmbt_irec *got,
> struct xfs_btree_cur *cur,
> int *logflags,
> @@ -5613,7 +5596,7 @@ xfs_bmap_shift_update_extent(
> *logflags |= XFS_ILOG_DEXT;
> }
>
> - xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork), idx, got);
> + xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork), ext, got);
>
> /* update reverse mapping */
> error = xfs_rmap_unmap_extent(mp, dfops, ip, whichfork, &prev);
> @@ -5638,7 +5621,7 @@ xfs_bmap_collapse_extents(
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> struct xfs_btree_cur *cur = NULL;
> struct xfs_bmbt_irec got, prev;
> - xfs_extnum_t current_ext;
> + struct xfs_iext_cursor ext;
> xfs_fileoff_t new_startoff;
> int error = 0;
> int logflags = 0;
> @@ -5669,14 +5652,14 @@ xfs_bmap_collapse_extents(
> cur->bc_private.b.flags = 0;
> }
>
> - if (!xfs_iext_lookup_extent(ip, ifp, *next_fsb, ¤t_ext, &got)) {
> + if (!xfs_iext_lookup_extent(ip, ifp, *next_fsb, &ext, &got)) {
> *done = true;
> goto del_cursor;
> }
> XFS_WANT_CORRUPTED_RETURN(mp, !isnullstartblock(got.br_startblock));
>
> new_startoff = got.br_startoff - offset_shift_fsb;
> - if (xfs_iext_get_extent(ifp, current_ext - 1, &prev)) {
> + if (xfs_iext_peek_prev_extent(ifp, &ext, &prev)) {
> if (new_startoff < prev.br_startoff + prev.br_blockcount) {
> error = -EINVAL;
> goto del_cursor;
> @@ -5684,8 +5667,8 @@ xfs_bmap_collapse_extents(
>
> if (xfs_bmse_can_merge(&prev, &got, offset_shift_fsb)) {
> error = xfs_bmse_merge(ip, whichfork, offset_shift_fsb,
> - ¤t_ext, &got, &prev, cur,
> - &logflags, dfops);
> + &ext, &got, &prev, cur, &logflags,
> + dfops);
> if (error)
> goto del_cursor;
> goto done;
> @@ -5697,13 +5680,13 @@ xfs_bmap_collapse_extents(
> }
> }
>
> - error = xfs_bmap_shift_update_extent(ip, whichfork, current_ext, &got,
> - cur, &logflags, dfops, new_startoff);
> + error = xfs_bmap_shift_update_extent(ip, whichfork, &ext, &got, cur,
> + &logflags, dfops, new_startoff);
> if (error)
> goto del_cursor;
>
> done:
> - if (!xfs_iext_get_extent(ifp, ++current_ext, &got)) {
> + if (!xfs_iext_next_extent(ifp, &ext, &got)) {
> *done = true;
> goto del_cursor;
> }
> @@ -5734,7 +5717,7 @@ xfs_bmap_insert_extents(
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> struct xfs_btree_cur *cur = NULL;
> struct xfs_bmbt_irec got, next;
> - xfs_extnum_t current_ext;
> + struct xfs_iext_cursor ext;
> xfs_fileoff_t new_startoff;
> int error = 0;
> int logflags = 0;
> @@ -5766,15 +5749,14 @@ xfs_bmap_insert_extents(
> }
>
> if (*next_fsb == NULLFSBLOCK) {
> - current_ext = xfs_iext_count(ifp) - 1;
> - if (!xfs_iext_get_extent(ifp, current_ext, &got) ||
> + xfs_iext_last(ifp, &ext);
> + if (!xfs_iext_get_extent(ifp, &ext, &got) ||
> stop_fsb > got.br_startoff) {
> *done = true;
> goto del_cursor;
> }
> } else {
> - if (!xfs_iext_lookup_extent(ip, ifp, *next_fsb, ¤t_ext,
> - &got)) {
> + if (!xfs_iext_lookup_extent(ip, ifp, *next_fsb, &ext, &got)) {
> *done = true;
> goto del_cursor;
> }
> @@ -5787,7 +5769,7 @@ xfs_bmap_insert_extents(
> }
>
> new_startoff = got.br_startoff + offset_shift_fsb;
> - if (xfs_iext_get_extent(ifp, current_ext + 1, &next)) {
> + if (xfs_iext_peek_next_extent(ifp, &ext, &next)) {
> if (new_startoff + got.br_blockcount > next.br_startoff) {
> error = -EINVAL;
> goto del_cursor;
> @@ -5803,12 +5785,12 @@ xfs_bmap_insert_extents(
> WARN_ON_ONCE(1);
> }
>
> - error = xfs_bmap_shift_update_extent(ip, whichfork, current_ext, &got,
> - cur, &logflags, dfops, new_startoff);
> + error = xfs_bmap_shift_update_extent(ip, whichfork, &ext, &got, cur,
> + &logflags, dfops, new_startoff);
> if (error)
> goto del_cursor;
>
> - if (!xfs_iext_get_extent(ifp, --current_ext, &got) ||
> + if (!xfs_iext_prev_extent(ifp, &ext, &got) ||
> stop_fsb >= got.br_startoff + got.br_blockcount) {
> *done = true;
> goto del_cursor;
> @@ -5825,10 +5807,10 @@ xfs_bmap_insert_extents(
> }
>
> /*
> - * Splits an extent into two extents at split_fsb block such that it is
> - * the first block of the current_ext. @current_ext is a target extent
> - * to be split. @split_fsb is a block where the extents is split.
> - * If split_fsb lies in a hole or the first block of extents, just return 0.
> + * Splits an extent into two extents at split_fsb block such that it is the
> + * first block of the current_ext. @ext is a target extent to be split.
> + * @split_fsb is a block where the extents is split. If split_fsb lies in a
> + * hole or the first block of extents, just return 0.
> */
> STATIC int
> xfs_bmap_split_extent_at(
> @@ -5845,7 +5827,7 @@ xfs_bmap_split_extent_at(
> struct xfs_mount *mp = ip->i_mount;
> struct xfs_ifork *ifp;
> xfs_fsblock_t gotblkcnt; /* new block count for got */
> - xfs_extnum_t current_ext;
> + struct xfs_iext_cursor ext;
> int error = 0;
> int logflags = 0;
> int i = 0;
> @@ -5873,7 +5855,7 @@ xfs_bmap_split_extent_at(
> /*
> * If there are not extents, or split_fsb lies in a hole we are done.
> */
> - if (!xfs_iext_lookup_extent(ip, ifp, split_fsb, ¤t_ext, &got) ||
> + if (!xfs_iext_lookup_extent(ip, ifp, split_fsb, &ext, &got) ||
> got.br_startoff >= split_fsb)
> return 0;
>
> @@ -5895,8 +5877,8 @@ xfs_bmap_split_extent_at(
> }
>
> got.br_blockcount = gotblkcnt;
> - xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork),
> - current_ext, &got);
> + xfs_iext_update_extent(ip, xfs_bmap_fork_to_state(whichfork), &ext,
> + &got);
>
> logflags = XFS_ILOG_CORE;
> if (cur) {
> @@ -5907,8 +5889,8 @@ xfs_bmap_split_extent_at(
> logflags |= XFS_ILOG_DEXT;
>
> /* Add new extent */
> - current_ext++;
> - xfs_iext_insert(ip, current_ext, 1, &new, 0);
> + xfs_iext_next(ifp, &ext);
> + xfs_iext_insert(ip, &ext, 1, &new, 0);
> XFS_IFORK_NEXT_SET(ip, whichfork,
> XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
>
> diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
> index a8777682ba57..195f335f4615 100644
> --- a/fs/xfs/libxfs/xfs_bmap.h
> +++ b/fs/xfs/libxfs/xfs_bmap.h
> @@ -43,7 +43,7 @@ struct xfs_bmalloca {
> xfs_fsblock_t blkno; /* starting block of new extent */
>
> struct xfs_btree_cur *cur; /* btree cursor */
> - xfs_extnum_t idx; /* current extent index */
> + struct xfs_iext_cursor ext;
> int nallocs;/* number of extents alloc'd */
> int logflags;/* flags for transaction logging */
>
> @@ -216,10 +216,11 @@ int xfs_bunmapi(struct xfs_trans *tp, struct xfs_inode *ip,
> xfs_extnum_t nexts, xfs_fsblock_t *firstblock,
> struct xfs_defer_ops *dfops, int *done);
> int xfs_bmap_del_extent_delay(struct xfs_inode *ip, int whichfork,
> - xfs_extnum_t *idx, struct xfs_bmbt_irec *got,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *got,
> + struct xfs_bmbt_irec *del);
> +void xfs_bmap_del_extent_cow(struct xfs_inode *ip,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *got,
> struct xfs_bmbt_irec *del);
> -void xfs_bmap_del_extent_cow(struct xfs_inode *ip, xfs_extnum_t *idx,
> - struct xfs_bmbt_irec *got, struct xfs_bmbt_irec *del);
> uint xfs_default_attroffset(struct xfs_inode *ip);
> int xfs_bmap_collapse_extents(struct xfs_trans *tp, struct xfs_inode *ip,
> xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb,
> @@ -232,7 +233,8 @@ int xfs_bmap_insert_extents(struct xfs_trans *tp, struct xfs_inode *ip,
> int xfs_bmap_split_extent(struct xfs_inode *ip, xfs_fileoff_t split_offset);
> int xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork,
> xfs_fileoff_t off, xfs_filblks_t len, xfs_filblks_t prealloc,
> - struct xfs_bmbt_irec *got, xfs_extnum_t *lastx, int eof);
> + struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur,
> + int eof);
>
> enum xfs_bmap_intent_type {
> XFS_BMAP_MAP = 1,
> diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
> index 7dd77b497fc2..c9e10d4818b7 100644
> --- a/fs/xfs/libxfs/xfs_inode_fork.c
> +++ b/fs/xfs/libxfs/xfs_inode_fork.c
> @@ -343,6 +343,7 @@ xfs_iformat_extents(
> int state = xfs_bmap_fork_to_state(whichfork);
> int nex = XFS_DFORK_NEXTENTS(dip, whichfork);
> int size = nex * sizeof(xfs_bmbt_rec_t);
> + struct xfs_iext_cursor ext;
> struct xfs_bmbt_rec *dp;
> int i;
>
> @@ -369,16 +370,21 @@ xfs_iformat_extents(
> ifp->if_bytes = size;
> if (size) {
> dp = (xfs_bmbt_rec_t *) XFS_DFORK_PTR(dip, whichfork);
> +
> + xfs_iext_first(ifp, &ext);
> for (i = 0; i < nex; i++, dp++) {
> xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
> +
> if (!xfs_bmbt_validate_extent(mp, whichfork, dp)) {
> XFS_ERROR_REPORT("xfs_iformat_extents(2)",
> XFS_ERRLEVEL_LOW, mp);
> return -EFSCORRUPTED;
> }
> +
> ep->l0 = get_unaligned_be64(&dp->l0);
> ep->l1 = get_unaligned_be64(&dp->l1);
> - trace_xfs_read_extent(ip, i, state, _THIS_IP_);
> + trace_xfs_read_extent(ip, &ext, state, _THIS_IP_);
> + xfs_iext_next(ifp, &ext);
> }
> }
> ifp->if_flags |= XFS_IFEXTENTS;
> @@ -739,17 +745,18 @@ xfs_iextents_copy(
> {
> int state = xfs_bmap_fork_to_state(whichfork);
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
> + struct xfs_iext_cursor ext;
> struct xfs_bmbt_irec rec;
> - int copied = 0, i = 0;
> + int copied = 0;
>
> ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED));
> ASSERT(ifp->if_bytes > 0);
>
> - while (xfs_iext_get_extent(ifp, i++, &rec)) {
> + for_each_iext(ifp, &ext, &rec) {
> if (isnullstartblock(rec.br_startblock))
> continue;
> xfs_bmbt_disk_set_all(dp, &rec);
> - trace_xfs_write_extent(ip, i, state, _RET_IP_);
> + trace_xfs_write_extent(ip, &ext, state, _RET_IP_);
> ASSERT(xfs_bmbt_validate_extent(ip->i_mount, whichfork, dp));
> copied += sizeof(struct xfs_bmbt_rec);
> dp++;
> @@ -894,7 +901,7 @@ xfs_iext_state_to_fork(
> void
> xfs_iext_insert(
> xfs_inode_t *ip, /* incore inode pointer */
> - xfs_extnum_t idx, /* starting index of new items */
> + struct xfs_iext_cursor *cur,
> xfs_extnum_t count, /* number of inserted items */
> xfs_bmbt_irec_t *new, /* items to insert */
> int state) /* type of extent conversion */
> @@ -902,12 +909,12 @@ xfs_iext_insert(
> xfs_ifork_t *ifp = xfs_iext_state_to_fork(ip, state);
> xfs_extnum_t i; /* extent record index */
>
> - trace_xfs_iext_insert(ip, idx, new, state, _RET_IP_);
> + trace_xfs_iext_insert(ip, cur->idx, new, state, _RET_IP_);
>
> ASSERT(ifp->if_flags & XFS_IFEXTENTS);
> - xfs_iext_add(ifp, idx, count);
> - for (i = idx; i < idx + count; i++, new++)
> - xfs_bmbt_set_all(xfs_iext_get_ext(ifp, i), new);
> + xfs_iext_add(ifp, cur->idx, count);
> + for (i = 0; i < count; i++, new++)
> + xfs_bmbt_set_all(xfs_iext_get_ext(ifp, cur->idx + i), new);
> }
>
> /*
> @@ -1145,7 +1152,7 @@ xfs_iext_add_indirect_multi(
> void
> xfs_iext_remove(
> xfs_inode_t *ip, /* incore inode pointer */
> - xfs_extnum_t idx, /* index to begin removing exts */
> + struct xfs_iext_cursor *cur,
> int ext_diff, /* number of extents to remove */
> int state) /* type of extent conversion */
> {
> @@ -1153,7 +1160,7 @@ xfs_iext_remove(
> xfs_extnum_t nextents; /* number of extents in file */
> int new_size; /* size of extents after removal */
>
> - trace_xfs_iext_remove(ip, idx, state, _RET_IP_);
> + trace_xfs_iext_remove(ip, cur, state, _RET_IP_);
>
> ASSERT(ext_diff > 0);
> nextents = xfs_iext_count(ifp);
> @@ -1162,11 +1169,11 @@ xfs_iext_remove(
> if (new_size == 0) {
> xfs_iext_destroy(ifp);
> } else if (ifp->if_flags & XFS_IFEXTIREC) {
> - xfs_iext_remove_indirect(ifp, idx, ext_diff);
> + xfs_iext_remove_indirect(ifp, cur->idx, ext_diff);
> } else if (ifp->if_real_bytes) {
> - xfs_iext_remove_direct(ifp, idx, ext_diff);
> + xfs_iext_remove_direct(ifp, cur->idx, ext_diff);
> } else {
> - xfs_iext_remove_inline(ifp, idx, ext_diff);
> + xfs_iext_remove_inline(ifp, cur->idx, ext_diff);
> }
> ifp->if_bytes = new_size;
> }
> @@ -1913,26 +1920,26 @@ xfs_ifork_init_cow(
> * Lookup the extent covering bno.
> *
> * If there is an extent covering bno return the extent index, and store the
> - * expanded extent structure in *gotp, and the extent index in *idx.
> + * expanded extent structure in *gotp, and the extent cursor in *cur.
> * If there is no extent covering bno, but there is an extent after it (e.g.
> - * it lies in a hole) return that extent in *gotp and its index in *idx
> + * it lies in a hole) return that extent in *gotp and its cursor in *cur
> * instead.
> - * If bno is beyond the last extent return false, and return the index after
> - * the last valid index in *idxp.
> + * If bno is beyond the last extent return false, and return an invalid
> + * cursor value.
> */
> bool
> xfs_iext_lookup_extent(
> struct xfs_inode *ip,
> struct xfs_ifork *ifp,
> xfs_fileoff_t bno,
> - xfs_extnum_t *idxp,
> + struct xfs_iext_cursor *cur,
> struct xfs_bmbt_irec *gotp)
> {
> struct xfs_bmbt_rec_host *ep;
>
> XFS_STATS_INC(ip->i_mount, xs_look_exlist);
>
> - ep = xfs_iext_bno_to_ext(ifp, bno, idxp);
> + ep = xfs_iext_bno_to_ext(ifp, bno, &cur->idx);
> if (!ep)
> return false;
> xfs_bmbt_get_all(ep, gotp);
> @@ -1948,31 +1955,31 @@ xfs_iext_lookup_extent_before(
> struct xfs_inode *ip,
> struct xfs_ifork *ifp,
> xfs_fileoff_t *end,
> - xfs_extnum_t *idxp,
> + struct xfs_iext_cursor *cur,
> struct xfs_bmbt_irec *gotp)
> {
> - if (xfs_iext_lookup_extent(ip, ifp, *end - 1, idxp, gotp) &&
> + if (xfs_iext_lookup_extent(ip, ifp, *end - 1, cur, gotp) &&
> gotp->br_startoff <= *end - 1)
> return true;
> - if (!xfs_iext_get_extent(ifp, --*idxp, gotp))
> + if (!xfs_iext_prev_extent(ifp, cur, gotp))
> return false;
> *end = gotp->br_startoff + gotp->br_blockcount;
> return true;
> }
>
> /*
> - * Return true if there is an extent at index idx, and return the expanded
> - * extent structure at idx in that case. Else return false.
> + * Return true if there is an extent at cursor cur and return the expanded
> + * extent structure at cur in gotp in that case. Else return false.
> */
> bool
> xfs_iext_get_extent(
> struct xfs_ifork *ifp,
> - xfs_extnum_t idx,
> + struct xfs_iext_cursor *cur,
> struct xfs_bmbt_irec *gotp)
> {
> - if (idx < 0 || idx >= xfs_iext_count(ifp))
> + if (cur->idx < 0 || cur->idx >= xfs_iext_count(ifp))
> return false;
> - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), gotp);
> + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, cur->idx), gotp);
> return true;
> }
>
> @@ -1980,15 +1987,15 @@ void
> xfs_iext_update_extent(
> struct xfs_inode *ip,
> int state,
> - xfs_extnum_t idx,
> + struct xfs_iext_cursor *cur,
> struct xfs_bmbt_irec *gotp)
> {
> struct xfs_ifork *ifp = xfs_iext_state_to_fork(ip, state);
>
> - ASSERT(idx >= 0);
> - ASSERT(idx < xfs_iext_count(ifp));
> + ASSERT(cur->idx >= 0);
> + ASSERT(cur->idx < xfs_iext_count(ifp));
>
> - trace_xfs_bmap_pre_update(ip, idx, state, _RET_IP_);
> - xfs_bmbt_set_all(xfs_iext_get_ext(ifp, idx), gotp);
> - trace_xfs_bmap_post_update(ip, idx, state, _RET_IP_);
> + trace_xfs_bmap_pre_update(ip, cur, state, _RET_IP_);
> + xfs_bmbt_set_all(xfs_iext_get_ext(ifp, cur->idx), gotp);
> + trace_xfs_bmap_post_update(ip, cur, state, _RET_IP_);
> }
> diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h
> index 113fd42ec36d..dc347dd9dc78 100644
> --- a/fs/xfs/libxfs/xfs_inode_fork.h
> +++ b/fs/xfs/libxfs/xfs_inode_fork.h
> @@ -151,12 +151,13 @@ void xfs_init_local_fork(struct xfs_inode *, int, const void *, int);
> struct xfs_bmbt_rec_host *
> xfs_iext_get_ext(struct xfs_ifork *, xfs_extnum_t);
> xfs_extnum_t xfs_iext_count(struct xfs_ifork *);
> -void xfs_iext_insert(struct xfs_inode *, xfs_extnum_t, xfs_extnum_t,
> - struct xfs_bmbt_irec *, int);
> +void xfs_iext_insert(struct xfs_inode *, struct xfs_iext_cursor *cur,
> + xfs_extnum_t, struct xfs_bmbt_irec *, int);
> void xfs_iext_add(struct xfs_ifork *, xfs_extnum_t, int);
> void xfs_iext_add_indirect_multi(struct xfs_ifork *, int,
> xfs_extnum_t, int);
> -void xfs_iext_remove(struct xfs_inode *, xfs_extnum_t, int, int);
> +void xfs_iext_remove(struct xfs_inode *, struct xfs_iext_cursor *,
> + int, int);
> void xfs_iext_remove_inline(struct xfs_ifork *, xfs_extnum_t, int);
> void xfs_iext_remove_direct(struct xfs_ifork *, xfs_extnum_t, int);
> void xfs_iext_remove_indirect(struct xfs_ifork *, xfs_extnum_t, int);
> @@ -182,15 +183,85 @@ void xfs_iext_irec_update_extoffs(struct xfs_ifork *, int, int);
>
> bool xfs_iext_lookup_extent(struct xfs_inode *ip,
> struct xfs_ifork *ifp, xfs_fileoff_t bno,
> - xfs_extnum_t *idxp, struct xfs_bmbt_irec *gotp);
> + struct xfs_iext_cursor *cur,
> + struct xfs_bmbt_irec *gotp);
> bool xfs_iext_lookup_extent_before(struct xfs_inode *ip,
> struct xfs_ifork *ifp, xfs_fileoff_t *end,
> - xfs_extnum_t *idxp, struct xfs_bmbt_irec *gotp);
> -
> -bool xfs_iext_get_extent(struct xfs_ifork *ifp, xfs_extnum_t idx,
> + struct xfs_iext_cursor *cur,
> + struct xfs_bmbt_irec *gotp);
> +bool xfs_iext_get_extent(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur,
> struct xfs_bmbt_irec *gotp);
> void xfs_iext_update_extent(struct xfs_inode *ip, int state,
> - xfs_extnum_t idx, struct xfs_bmbt_irec *gotp);
> + struct xfs_iext_cursor *cur,
> + struct xfs_bmbt_irec *gotp);
> +
> +static inline void xfs_iext_first(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur)
> +{
> + cur->idx = 0;
> +}
> +
> +static inline void xfs_iext_last(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur)
> +{
> + cur->idx = xfs_iext_count(ifp) - 1;
> +}
> +
> +static inline void xfs_iext_next(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur)
> +{
> + cur->idx++;
> +}
> +
> +static inline void xfs_iext_prev(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur)
> +{
> + cur->idx--;
> +}
> +
> +static inline bool xfs_iext_next_extent(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
> +{
> + xfs_iext_next(ifp, cur);
> + return xfs_iext_get_extent(ifp, cur, gotp);
> +}
> +
> +static inline bool xfs_iext_prev_extent(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
> +{
> + xfs_iext_prev(ifp, cur);
> + return xfs_iext_get_extent(ifp, cur, gotp);
> +}
> +
> +/*
> + * Return the extent after cur in gotp without updating the cursor.
> + */
> +static inline bool xfs_iext_peek_next_extent(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
> +{
> + struct xfs_iext_cursor ncur = *cur;
> +
> + xfs_iext_next(ifp, &ncur);
> + return xfs_iext_get_extent(ifp, &ncur, gotp);
> +}
> +
> +/*
> + * Return the extent before cur in gotp without updating the cursor.
> + */
> +static inline bool xfs_iext_peek_prev_extent(struct xfs_ifork *ifp,
> + struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
> +{
> + struct xfs_iext_cursor ncur = *cur;
> +
> + xfs_iext_prev(ifp, &ncur);
> + return xfs_iext_get_extent(ifp, &ncur, gotp);
> +}
> +
> +#define for_each_iext(ifp, ext, got) \
> + for (xfs_iext_first((ifp), (ext)); \
> + xfs_iext_get_extent((ifp), (ext), (got)); \
> + xfs_iext_next((ifp), (ext)))
>
> extern struct kmem_zone *xfs_ifork_zone;
>
> diff --git a/fs/xfs/libxfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h
> index f04dbfb2f50d..5da6382bdaf1 100644
> --- a/fs/xfs/libxfs/xfs_types.h
> +++ b/fs/xfs/libxfs/xfs_types.h
> @@ -142,5 +142,8 @@ typedef uint32_t xfs_dqid_t;
> #define XFS_NBWORD (1 << XFS_NBWORDLOG)
> #define XFS_WORDMASK ((1 << XFS_WORDLOG) - 1)
>
> +struct xfs_iext_cursor {
> + xfs_extnum_t idx;
> +};
>
> #endif /* __XFS_TYPES_H__ */
> diff --git a/fs/xfs/scrub/bmap.c b/fs/xfs/scrub/bmap.c
> index 3c17b182616f..778b709dbd0c 100644
> --- a/fs/xfs/scrub/bmap.c
> +++ b/fs/xfs/scrub/bmap.c
> @@ -237,7 +237,7 @@ xfs_scrub_bmap(
> struct xfs_inode *ip = sc->ip;
> struct xfs_ifork *ifp;
> xfs_fileoff_t endoff;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> bool found;
> int error = 0;
>
> @@ -317,9 +317,9 @@ xfs_scrub_bmap(
> /* Scrub extent records. */
> info.lastoff = 0;
> ifp = XFS_IFORK_PTR(ip, whichfork);
> - for (found = xfs_iext_lookup_extent(ip, ifp, 0, &idx, &irec);
> + for (found = xfs_iext_lookup_extent(ip, ifp, 0, &ext, &irec);
> found != 0;
> - found = xfs_iext_get_extent(ifp, ++idx, &irec)) {
> + found = xfs_iext_next_extent(ifp, &ext, &irec)) {
> if (xfs_scrub_should_terminate(sc, &error))
> break;
> if (isnullstartblock(irec.br_startblock))
> diff --git a/fs/xfs/scrub/dir.c b/fs/xfs/scrub/dir.c
> index 169fb10daaaa..46765102638c 100644
> --- a/fs/xfs/scrub/dir.c
> +++ b/fs/xfs/scrub/dir.c
> @@ -614,7 +614,7 @@ xfs_scrub_directory_blocks(
> xfs_fileoff_t leaf_lblk;
> xfs_fileoff_t free_lblk;
> xfs_fileoff_t lblk;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> xfs_dablk_t dabno;
> bool found;
> int is_block = 0;
> @@ -639,7 +639,7 @@ xfs_scrub_directory_blocks(
> goto out;
>
> /* Iterate all the data extents in the directory... */
> - found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &idx, &got);
> + found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &ext, &got);
> while (found) {
> /* Block directories only have a single block at offset 0. */
> if (is_block &&
> @@ -676,17 +676,17 @@ xfs_scrub_directory_blocks(
> }
> dabno = got.br_startoff + got.br_blockcount;
> lblk = roundup(dabno, args.geo->fsbcount);
> - found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &idx, &got);
> + found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &ext, &got);
> }
>
> if (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT)
> goto out;
>
> /* Look for a leaf1 block, which has free info. */
> - if (xfs_iext_lookup_extent(sc->ip, ifp, leaf_lblk, &idx, &got) &&
> + if (xfs_iext_lookup_extent(sc->ip, ifp, leaf_lblk, &ext, &got) &&
> got.br_startoff == leaf_lblk &&
> got.br_blockcount == args.geo->fsbcount &&
> - !xfs_iext_get_extent(ifp, ++idx, &got)) {
> + !xfs_iext_next_extent(ifp, &ext, &got)) {
> if (is_block) {
> xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, lblk);
> goto out;
> @@ -702,7 +702,7 @@ xfs_scrub_directory_blocks(
>
> /* Scan for free blocks */
> lblk = free_lblk;
> - found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &idx, &got);
> + found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &ext, &got);
> while (found) {
> /*
> * Dirs can't have blocks mapped above 2^32.
> @@ -740,7 +740,7 @@ xfs_scrub_directory_blocks(
> }
> dabno = got.br_startoff + got.br_blockcount;
> lblk = roundup(dabno, args.geo->fsbcount);
> - found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &idx, &got);
> + found = xfs_iext_lookup_extent(sc->ip, ifp, lblk, &ext, &got);
> }
> out:
> return error;
> diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
> index 170b74c7f2d5..b6b954d5cf54 100644
> --- a/fs/xfs/xfs_bmap_util.c
> +++ b/fs/xfs/xfs_bmap_util.c
> @@ -229,15 +229,17 @@ xfs_bmap_count_leaves(
> struct xfs_ifork *ifp,
> xfs_filblks_t *count)
> {
> + struct xfs_iext_cursor ext;
> struct xfs_bmbt_irec got;
> - xfs_extnum_t numrecs = 0, i = 0;
> + xfs_extnum_t numrecs = 0;
>
> - while (xfs_iext_get_extent(ifp, i++, &got)) {
> + for_each_iext(ifp, &ext, &got) {
> if (!isnullstartblock(got.br_startblock)) {
> *count += got.br_blockcount;
> numrecs++;
> }
> }
> +
> return numrecs;
> }
>
> @@ -525,7 +527,7 @@ xfs_getbmap(
> struct xfs_ifork *ifp;
> struct xfs_bmbt_irec got, rec;
> xfs_filblks_t len;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> if (bmv->bmv_iflags & ~BMV_IF_VALID)
> return -EINVAL;
> @@ -629,7 +631,7 @@ xfs_getbmap(
> goto out_unlock_ilock;
> }
>
> - if (!xfs_iext_lookup_extent(ip, ifp, bno, &idx, &got)) {
> + if (!xfs_iext_lookup_extent(ip, ifp, bno, &ext, &got)) {
> /*
> * Report a whole-file hole if the delalloc flag is set to
> * stay compatible with the old implementation.
> @@ -668,7 +670,7 @@ xfs_getbmap(
> goto out_unlock_ilock;
> } while (xfs_getbmap_next_rec(&rec, bno));
>
> - if (!xfs_iext_get_extent(ifp, ++idx, &got)) {
> + if (!xfs_iext_next_extent(ifp, &ext, &got)) {
> xfs_fileoff_t end = XFS_B_TO_FSB(mp, XFS_ISIZE(ip));
>
> out[bmv->bmv_entries - 1].bmv_oflags |= BMV_OF_LAST;
> diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
> index 238e3650a9d2..ad54fd775bda 100644
> --- a/fs/xfs/xfs_dir2_readdir.c
> +++ b/fs/xfs/xfs_dir2_readdir.c
> @@ -266,7 +266,7 @@ xfs_dir2_leaf_readbuf(
> xfs_dablk_t next_ra;
> xfs_dablk_t map_off;
> xfs_dablk_t last_da;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> int ra_want;
> int error = 0;
>
> @@ -283,7 +283,7 @@ xfs_dir2_leaf_readbuf(
> */
> last_da = xfs_dir2_byte_to_da(geo, XFS_DIR2_LEAF_OFFSET);
> map_off = xfs_dir2_db_to_da(geo, xfs_dir2_byte_to_db(geo, *cur_off));
> - if (!xfs_iext_lookup_extent(dp, ifp, map_off, &idx, &map))
> + if (!xfs_iext_lookup_extent(dp, ifp, map_off, &ext, &map))
> goto out;
> if (map.br_startoff >= last_da)
> goto out;
> @@ -311,7 +311,7 @@ xfs_dir2_leaf_readbuf(
> if (next_ra >= last_da)
> goto out_no_ra;
> if (map.br_blockcount < geo->fsbcount &&
> - !xfs_iext_get_extent(ifp, ++idx, &map))
> + !xfs_iext_next_extent(ifp, &ext, &map))
> goto out_no_ra;
> if (map.br_startoff >= last_da)
> goto out_no_ra;
> @@ -334,7 +334,7 @@ xfs_dir2_leaf_readbuf(
> ra_want -= geo->fsbcount;
> next_ra += geo->fsbcount;
> }
> - if (!xfs_iext_get_extent(ifp, ++idx, &map)) {
> + if (!xfs_iext_next_extent(ifp, &ext, &map)) {
> *ra_blk = last_da;
> break;
> }
> diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
> index cd82429d8df7..8338b894d54f 100644
> --- a/fs/xfs/xfs_dquot.c
> +++ b/fs/xfs/xfs_dquot.c
> @@ -703,7 +703,7 @@ xfs_dq_get_next_id(
> xfs_dqid_t next_id = *id + 1; /* simple advance */
> uint lock_flags;
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor cur;
> xfs_fsblock_t start;
> int error = 0;
>
> @@ -727,7 +727,7 @@ xfs_dq_get_next_id(
> return error;
> }
>
> - if (xfs_iext_lookup_extent(quotip, "ip->i_df, start, &idx, &got)) {
> + if (xfs_iext_lookup_extent(quotip, "ip->i_df, start, &cur, &got)) {
> /* contiguous chunk, bump startoff for the id calculation */
> if (got.br_startoff < start)
> got.br_startoff = start;
> diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
> index f179bdf1644d..046ade883611 100644
> --- a/fs/xfs/xfs_iomap.c
> +++ b/fs/xfs/xfs_iomap.c
> @@ -389,7 +389,7 @@ xfs_iomap_prealloc_size(
> struct xfs_inode *ip,
> loff_t offset,
> loff_t count,
> - xfs_extnum_t idx)
> + struct xfs_iext_cursor *ext)
> {
> struct xfs_mount *mp = ip->i_mount;
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
> @@ -414,7 +414,7 @@ xfs_iomap_prealloc_size(
> */
> if ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ||
> XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_dalign) ||
> - !xfs_iext_get_extent(ifp, idx - 1, &prev) ||
> + !xfs_iext_peek_prev_extent(ifp, ext, &prev) ||
> prev.br_startoff + prev.br_blockcount < offset_fsb)
> return mp->m_writeio_blocks;
>
> @@ -532,7 +532,7 @@ xfs_file_iomap_begin_delay(
> xfs_fileoff_t end_fsb;
> int error = 0, eof = 0;
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> xfs_fsblock_t prealloc_blocks = 0;
>
> ASSERT(!XFS_IS_REALTIME_INODE(ip));
> @@ -557,7 +557,7 @@ xfs_file_iomap_begin_delay(
> goto out_unlock;
> }
>
> - eof = !xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got);
> + eof = !xfs_iext_lookup_extent(ip, ifp, offset_fsb, &ext, &got);
> if (!eof && got.br_startoff <= offset_fsb) {
> if (xfs_is_reflink_inode(ip)) {
> bool shared;
> @@ -591,7 +591,8 @@ xfs_file_iomap_begin_delay(
> end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
>
> if (eof) {
> - prealloc_blocks = xfs_iomap_prealloc_size(ip, offset, count, idx);
> + prealloc_blocks = xfs_iomap_prealloc_size(ip, offset, count,
> + &ext);
> if (prealloc_blocks) {
> xfs_extlen_t align;
> xfs_off_t end_offset;
> @@ -613,7 +614,7 @@ xfs_file_iomap_begin_delay(
>
> retry:
> error = xfs_bmapi_reserve_delalloc(ip, XFS_DATA_FORK, offset_fsb,
> - end_fsb - offset_fsb, prealloc_blocks, &got, &idx, eof);
> + end_fsb - offset_fsb, prealloc_blocks, &got, &ext, eof);
> switch (error) {
> case 0:
> break;
> diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c
> index 1205747e1409..cadf0ff68003 100644
> --- a/fs/xfs/xfs_reflink.c
> +++ b/fs/xfs/xfs_reflink.c
> @@ -273,7 +273,7 @@ xfs_reflink_reserve_cow(
> struct xfs_bmbt_irec got;
> int error = 0;
> bool eof = false, trimmed;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> /*
> * Search the COW fork extent list first. This serves two purposes:
> @@ -284,7 +284,7 @@ xfs_reflink_reserve_cow(
> * tree.
> */
>
> - if (!xfs_iext_lookup_extent(ip, ifp, imap->br_startoff, &idx, &got))
> + if (!xfs_iext_lookup_extent(ip, ifp, imap->br_startoff, &ext, &got))
> eof = true;
> if (!eof && got.br_startoff <= imap->br_startoff) {
> trace_xfs_reflink_cow_found(ip, imap);
> @@ -312,7 +312,7 @@ xfs_reflink_reserve_cow(
> return error;
>
> error = xfs_bmapi_reserve_delalloc(ip, XFS_COW_FORK, imap->br_startoff,
> - imap->br_blockcount, 0, &got, &idx, eof);
> + imap->br_blockcount, 0, &got, &ext, eof);
> if (error == -ENOSPC || error == -EDQUOT)
> trace_xfs_reflink_cow_enospc(ip, imap);
> if (error)
> @@ -359,16 +359,16 @@ xfs_reflink_convert_cow(
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK);
> xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset);
> xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + count);
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> bool found;
> int error = 0;
>
> xfs_ilock(ip, XFS_ILOCK_EXCL);
>
> /* Convert all the extents to real from unwritten. */
> - for (found = xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got);
> + for (found = xfs_iext_lookup_extent(ip, ifp, offset_fsb, &ext, &got);
> found && got.br_startoff < end_fsb;
> - found = xfs_iext_get_extent(ifp, ++idx, &got)) {
> + found = xfs_iext_next_extent(ifp, &ext, &got)) {
> error = xfs_reflink_convert_cow_extent(ip, &got, offset_fsb,
> end_fsb - offset_fsb, &dfops);
> if (error)
> @@ -399,7 +399,7 @@ xfs_reflink_allocate_cow(
> bool trimmed;
> xfs_filblks_t resaligned;
> xfs_extlen_t resblks = 0;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> retry:
> ASSERT(xfs_is_reflink_inode(ip));
> @@ -409,7 +409,7 @@ xfs_reflink_allocate_cow(
> * Even if the extent is not shared we might have a preallocation for
> * it in the COW fork. If so use it.
> */
> - if (xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &idx, &got) &&
> + if (xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &ext, &got) &&
> got.br_startoff <= offset_fsb) {
> *shared = true;
>
> @@ -496,13 +496,13 @@ xfs_reflink_find_cow_mapping(
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK);
> xfs_fileoff_t offset_fsb;
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL | XFS_ILOCK_SHARED));
> ASSERT(xfs_is_reflink_inode(ip));
>
> offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset);
> - if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got))
> + if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &ext, &got))
> return false;
> if (got.br_startoff > offset_fsb)
> return false;
> @@ -524,18 +524,18 @@ xfs_reflink_trim_irec_to_next_cow(
> {
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK);
> struct xfs_bmbt_irec got;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> if (!xfs_is_reflink_inode(ip))
> return;
>
> /* Find the extent in the CoW fork. */
> - if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got))
> + if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &ext, &got))
> return;
>
> /* This is the extent before; try sliding up one. */
> if (got.br_startoff < offset_fsb) {
> - if (!xfs_iext_get_extent(ifp, idx + 1, &got))
> + if (!xfs_iext_next_extent(ifp, &ext, &got))
> return;
> }
>
> @@ -562,14 +562,14 @@ xfs_reflink_cancel_cow_blocks(
> {
> struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_COW_FORK);
> struct xfs_bmbt_irec got, del;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> xfs_fsblock_t firstfsb;
> struct xfs_defer_ops dfops;
> int error = 0;
>
> if (!xfs_is_reflink_inode(ip))
> return 0;
> - if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &idx, &got))
> + if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &ext, &got))
> return 0;
>
> while (got.br_startoff < end_fsb) {
> @@ -579,7 +579,7 @@ xfs_reflink_cancel_cow_blocks(
>
> if (isnullstartblock(del.br_startblock)) {
> error = xfs_bmap_del_extent_delay(ip, XFS_COW_FORK,
> - &idx, &got, &del);
> + &ext, &got, &del);
> if (error)
> break;
> } else if (del.br_state == XFS_EXT_UNWRITTEN || cancel_real) {
> @@ -610,10 +610,10 @@ xfs_reflink_cancel_cow_blocks(
> }
>
> /* Remove the mapping from the CoW fork. */
> - xfs_bmap_del_extent_cow(ip, &idx, &got, &del);
> + xfs_bmap_del_extent_cow(ip, &ext, &got, &del);
> }
>
> - if (!xfs_iext_get_extent(ifp, ++idx, &got))
> + if (!xfs_iext_next_extent(ifp, &ext, &got))
> break;
> }
>
> @@ -698,7 +698,7 @@ xfs_reflink_end_cow(
> int error;
> unsigned int resblks;
> xfs_filblks_t rlen;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
>
> trace_xfs_reflink_end_cow(ip, offset, count);
>
> @@ -738,7 +738,7 @@ xfs_reflink_end_cow(
> * left by the time I/O completes for the loser of the race. In that
> * case we are done.
> */
> - if (!xfs_iext_lookup_extent_before(ip, ifp, &end_fsb, &idx, &got))
> + if (!xfs_iext_lookup_extent_before(ip, ifp, &end_fsb, &ext, &got))
> goto out_cancel;
>
> /* Walk backwards until we're out of the I/O range... */
> @@ -746,9 +746,9 @@ xfs_reflink_end_cow(
> del = got;
> xfs_trim_extent(&del, offset_fsb, end_fsb - offset_fsb);
>
> - /* Extent delete may have bumped idx forward */
> + /* Extent delete may have bumped ext forward */
> if (!del.br_blockcount) {
> - idx--;
> + xfs_iext_prev(ifp, &ext);
> goto next_extent;
> }
>
> @@ -760,7 +760,7 @@ xfs_reflink_end_cow(
> * allocated but have not yet been involved in a write.
> */
> if (got.br_state == XFS_EXT_UNWRITTEN) {
> - idx--;
> + xfs_iext_prev(ifp, &ext);
> goto next_extent;
> }
>
> @@ -791,14 +791,14 @@ xfs_reflink_end_cow(
> goto out_defer;
>
> /* Remove the mapping from the CoW fork. */
> - xfs_bmap_del_extent_cow(ip, &idx, &got, &del);
> + xfs_bmap_del_extent_cow(ip, &ext, &got, &del);
>
> xfs_defer_ijoin(&dfops, ip);
> error = xfs_defer_finish(&tp, &dfops);
> if (error)
> goto out_defer;
> next_extent:
> - if (!xfs_iext_get_extent(ifp, idx, &got))
> + if (!xfs_iext_get_extent(ifp, &ext, &got))
> break;
> }
>
> @@ -1428,7 +1428,7 @@ xfs_reflink_inode_has_shared_extents(
> xfs_extlen_t aglen;
> xfs_agblock_t rbno;
> xfs_extlen_t rlen;
> - xfs_extnum_t idx;
> + struct xfs_iext_cursor ext;
> bool found;
> int error;
>
> @@ -1440,7 +1440,7 @@ xfs_reflink_inode_has_shared_extents(
> }
>
> *has_shared = false;
> - found = xfs_iext_lookup_extent(ip, ifp, 0, &idx, &got);
> + found = xfs_iext_lookup_extent(ip, ifp, 0, &ext, &got);
> while (found) {
> if (isnullstartblock(got.br_startblock) ||
> got.br_state != XFS_EXT_NORM)
> @@ -1459,7 +1459,7 @@ xfs_reflink_inode_has_shared_extents(
> return 0;
> }
> next:
> - found = xfs_iext_get_extent(ifp, ++idx, &got);
> + found = xfs_iext_next_extent(ifp, &ext, &got);
> }
>
> return 0;
> diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
> index 665ef6cca90c..667bfce802cd 100644
> --- a/fs/xfs/xfs_trace.h
> +++ b/fs/xfs/xfs_trace.h
> @@ -258,9 +258,9 @@ TRACE_EVENT(xfs_iext_insert,
> );
>
> DECLARE_EVENT_CLASS(xfs_bmap_class,
> - TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state,
> + TP_PROTO(struct xfs_inode *ip, struct xfs_iext_cursor *cur, int state,
> unsigned long caller_ip),
> - TP_ARGS(ip, idx, state, caller_ip),
> + TP_ARGS(ip, cur, state, caller_ip),
> TP_STRUCT__entry(
> __field(dev_t, dev)
> __field(xfs_ino_t, ino)
> @@ -277,10 +277,10 @@ DECLARE_EVENT_CLASS(xfs_bmap_class,
> struct xfs_bmbt_irec r;
>
> ifp = xfs_iext_state_to_fork(ip, state);
> - xfs_iext_get_extent(ifp, idx, &r);
> + xfs_iext_get_extent(ifp, cur, &r);
> __entry->dev = VFS_I(ip)->i_sb->s_dev;
> __entry->ino = ip->i_ino;
> - __entry->idx = idx;
> + __entry->idx = cur->idx;
> __entry->startoff = r.br_startoff;
> __entry->startblock = r.br_startblock;
> __entry->blockcount = r.br_blockcount;
> @@ -303,9 +303,9 @@ DECLARE_EVENT_CLASS(xfs_bmap_class,
>
> #define DEFINE_BMAP_EVENT(name) \
> DEFINE_EVENT(xfs_bmap_class, name, \
> - TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, \
> + TP_PROTO(struct xfs_inode *ip, struct xfs_iext_cursor *cur, int state, \
> unsigned long caller_ip), \
> - TP_ARGS(ip, idx, state, caller_ip))
> + TP_ARGS(ip, cur, state, caller_ip))
> DEFINE_BMAP_EVENT(xfs_iext_remove);
> DEFINE_BMAP_EVENT(xfs_bmap_pre_update);
> DEFINE_BMAP_EVENT(xfs_bmap_post_update);
> --
> 2.14.2
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2017-10-31 22:02 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-31 14:22 b+tree for the incore extent list Christoph Hellwig
2017-10-31 14:22 ` [PATCH 01/18] xfs: pass an on-disk extent to xfs_bmbt_validate_extent Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:15 ` Darrick J. Wong
2017-11-01 13:58 ` Brian Foster
2017-11-01 23:00 ` Darrick J. Wong
2017-11-02 11:57 ` Brian Foster
2017-11-02 16:05 ` Darrick J. Wong
2017-11-02 16:54 ` Brian Foster
2017-11-02 18:42 ` Christoph Hellwig
2017-11-02 19:35 ` Brian Foster
2017-11-02 23:45 ` Dave Chinner
2017-10-31 14:22 ` [PATCH 02/18] xfs: don't create overlapping extents in xfs_bmap_add_extent_delay_real Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:34 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 03/18] xfs: treat idx as a cursor " Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:35 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 04/18] xfs: treat idx as a cursor in xfs_bmap_add_extent_hole_delay Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:35 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 05/18] xfs: treat idx as a cursor in xfs_bmap_add_extent_hole_real Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:35 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 06/18] xfs: treat idx as a cursor in xfs_bmap_add_extent_unwritten_real Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:36 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 07/18] xfs: treat idx as a cursor in xfs_bmap_del_extent_* Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:37 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 08/18] xfs: treat idx as a cursor in xfs_bmap_collapse_extents Christoph Hellwig
2017-10-31 17:53 ` Brian Foster
2017-10-31 21:37 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 09/18] xfs: allow unaligned extent records in xfs_bmbt_disk_set_all Christoph Hellwig
2017-10-31 21:34 ` Darrick J. Wong
2017-11-02 13:54 ` Brian Foster
2017-10-31 14:22 ` [PATCH 10/18] xfs: iterate over extents in xfs_iextents_copy Christoph Hellwig
2017-10-31 21:41 ` Darrick J. Wong
2017-11-02 13:54 ` Brian Foster
2017-10-31 14:22 ` [PATCH 11/18] xfs: iterate over extents in xfs_bmap_extents_to_btree Christoph Hellwig
2017-10-31 21:41 ` Darrick J. Wong
2017-11-02 13:54 ` Brian Foster
2017-10-31 14:22 ` [PATCH 12/18] xfs: introduce the xfs_iext_cursor abstraction Christoph Hellwig
2017-10-31 22:02 ` Darrick J. Wong [this message]
2017-11-02 18:49 ` Christoph Hellwig
2017-11-02 19:01 ` Darrick J. Wong
2017-11-02 19:11 ` Christoph Hellwig
2017-11-02 17:14 ` Brian Foster
2017-11-02 18:51 ` Christoph Hellwig
2017-11-02 19:36 ` Brian Foster
2017-11-03 7:26 ` Christoph Hellwig
2017-10-31 14:22 ` [PATCH 13/18] xfs: iterate backwards in xfs_reflink_cancel_cow_blocks Christoph Hellwig
2017-10-31 22:10 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 14/18] xfs: simplify xfs_reflink_convert_cow Christoph Hellwig
2017-10-31 22:20 ` Darrick J. Wong
2017-11-02 18:56 ` Christoph Hellwig
2017-10-31 14:22 ` [PATCH 15/18] xfs: remove support for inlining data/extents into the inode fork Christoph Hellwig
2017-10-31 22:35 ` Darrick J. Wong
2017-11-02 18:57 ` Christoph Hellwig
2017-11-02 19:26 ` Darrick J. Wong
2017-11-02 21:43 ` Dave Chinner
2017-11-02 22:08 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 16/18] xfs: use a b+tree for the in-core extent list Christoph Hellwig
2017-11-01 18:47 ` Darrick J. Wong
2017-11-02 0:16 ` Darrick J. Wong
2017-11-02 6:03 ` Christoph Hellwig
2017-11-02 0:14 ` Darrick J. Wong
2017-11-02 19:09 ` Christoph Hellwig
2017-10-31 14:22 ` [PATCH 17/18] xfs: remove the nr_extents argument to xfs_iext_insert Christoph Hellwig
2017-10-31 22:35 ` Darrick J. Wong
2017-10-31 14:22 ` [PATCH 18/18] xfs: remove the nr_extents argument to xfs_iext_remove Christoph Hellwig
2017-10-31 22:37 ` Darrick J. Wong
2017-11-01 3:08 ` b+tree for the incore extent list Darrick J. Wong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20171031220234.GT4911@magnolia \
--to=darrick.wong@oracle.com \
--cc=hch@lst.de \
--cc=linux-xfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).