diff options
Diffstat (limited to 'fs/xfs/xfs_dir2_node.c')
-rw-r--r-- | fs/xfs/xfs_dir2_node.c | 1340 |
1 files changed, 827 insertions, 513 deletions
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index b0f26780449d..2226a00acd15 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -1,5 +1,6 @@ /* * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2013 Red Hat, Inc. * All Rights Reserved. * * This program is free software; you can redistribute it and/or @@ -32,47 +33,275 @@ #include "xfs_dir2_priv.h" #include "xfs_error.h" #include "xfs_trace.h" +#include "xfs_buf_item.h" +#include "xfs_cksum.h" /* * Function declarations. */ -static void xfs_dir2_free_log_header(xfs_trans_t *tp, xfs_dabuf_t *bp); -static int xfs_dir2_leafn_add(xfs_dabuf_t *bp, xfs_da_args_t *args, int index); -#ifdef DEBUG -static void xfs_dir2_leafn_check(xfs_inode_t *dp, xfs_dabuf_t *bp); -#else -#define xfs_dir2_leafn_check(dp, bp) -#endif -static void xfs_dir2_leafn_moveents(xfs_da_args_t *args, xfs_dabuf_t *bp_s, - int start_s, xfs_dabuf_t *bp_d, int start_d, - int count); +static int xfs_dir2_leafn_add(struct xfs_buf *bp, xfs_da_args_t *args, + int index); static void xfs_dir2_leafn_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1, xfs_da_state_blk_t *blk2); -static int xfs_dir2_leafn_remove(xfs_da_args_t *args, xfs_dabuf_t *bp, +static int xfs_dir2_leafn_remove(xfs_da_args_t *args, struct xfs_buf *bp, int index, xfs_da_state_blk_t *dblk, int *rval); static int xfs_dir2_node_addname_int(xfs_da_args_t *args, xfs_da_state_blk_t *fblk); /* + * Check internal consistency of a leafn block. + */ +#ifdef DEBUG +#define xfs_dir3_leaf_check(mp, bp) \ +do { \ + if (!xfs_dir3_leafn_check((mp), (bp))) \ + ASSERT(0); \ +} while (0); + +static bool +xfs_dir3_leafn_check( + struct xfs_mount *mp, + struct xfs_buf *bp) +{ + struct xfs_dir2_leaf *leaf = bp->b_addr; + struct xfs_dir3_icleaf_hdr leafhdr; + + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + + if (leafhdr.magic == XFS_DIR3_LEAFN_MAGIC) { + struct xfs_dir3_leaf_hdr *leaf3 = bp->b_addr; + if (be64_to_cpu(leaf3->info.blkno) != bp->b_bn) + return false; + } else if (leafhdr.magic != XFS_DIR2_LEAFN_MAGIC) + return false; + + return xfs_dir3_leaf_check_int(mp, &leafhdr, leaf); +} +#else +#define xfs_dir3_leaf_check(mp, bp) +#endif + +static bool +xfs_dir3_free_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_dir2_free_hdr *hdr = bp->b_addr; + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr; + + if (hdr3->magic != cpu_to_be32(XFS_DIR3_FREE_MAGIC)) + return false; + if (!uuid_equal(&hdr3->uuid, &mp->m_sb.sb_uuid)) + return false; + if (be64_to_cpu(hdr3->blkno) != bp->b_bn) + return false; + } else { + if (hdr->magic != cpu_to_be32(XFS_DIR2_FREE_MAGIC)) + return false; + } + + /* XXX: should bounds check the xfs_dir3_icfree_hdr here */ + + return true; +} + +static void +xfs_dir3_free_read_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + + if ((xfs_sb_version_hascrc(&mp->m_sb) && + !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length), + XFS_DIR3_FREE_CRC_OFF)) || + !xfs_dir3_free_verify(bp)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); + xfs_buf_ioerror(bp, EFSCORRUPTED); + } +} + +static void +xfs_dir3_free_write_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_buf_log_item *bip = bp->b_fspriv; + struct xfs_dir3_blk_hdr *hdr3 = bp->b_addr; + + if (!xfs_dir3_free_verify(bp)) { + XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr); + xfs_buf_ioerror(bp, EFSCORRUPTED); + return; + } + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return; + + if (bip) + hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn); + + xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_FREE_CRC_OFF); +} + +const struct xfs_buf_ops xfs_dir3_free_buf_ops = { + .verify_read = xfs_dir3_free_read_verify, + .verify_write = xfs_dir3_free_write_verify, +}; + + +static int +__xfs_dir3_free_read( + struct xfs_trans *tp, + struct xfs_inode *dp, + xfs_dablk_t fbno, + xfs_daddr_t mappedbno, + struct xfs_buf **bpp) +{ + int err; + + err = xfs_da_read_buf(tp, dp, fbno, mappedbno, bpp, + XFS_DATA_FORK, &xfs_dir3_free_buf_ops); + + /* try read returns without an error or *bpp if it lands in a hole */ + if (!err && tp && *bpp) + xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_FREE_BUF); + return err; +} + +int +xfs_dir2_free_read( + struct xfs_trans *tp, + struct xfs_inode *dp, + xfs_dablk_t fbno, + struct xfs_buf **bpp) +{ + return __xfs_dir3_free_read(tp, dp, fbno, -1, bpp); +} + +static int +xfs_dir2_free_try_read( + struct xfs_trans *tp, + struct xfs_inode *dp, + xfs_dablk_t fbno, + struct xfs_buf **bpp) +{ + return __xfs_dir3_free_read(tp, dp, fbno, -2, bpp); +} + + +void +xfs_dir3_free_hdr_from_disk( + struct xfs_dir3_icfree_hdr *to, + struct xfs_dir2_free *from) +{ + if (from->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)) { + to->magic = be32_to_cpu(from->hdr.magic); + to->firstdb = be32_to_cpu(from->hdr.firstdb); + to->nvalid = be32_to_cpu(from->hdr.nvalid); + to->nused = be32_to_cpu(from->hdr.nused); + } else { + struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)from; + + to->magic = be32_to_cpu(hdr3->hdr.magic); + to->firstdb = be32_to_cpu(hdr3->firstdb); + to->nvalid = be32_to_cpu(hdr3->nvalid); + to->nused = be32_to_cpu(hdr3->nused); + } + + ASSERT(to->magic == XFS_DIR2_FREE_MAGIC || + to->magic == XFS_DIR3_FREE_MAGIC); +} + +static void +xfs_dir3_free_hdr_to_disk( + struct xfs_dir2_free *to, + struct xfs_dir3_icfree_hdr *from) +{ + ASSERT(from->magic == XFS_DIR2_FREE_MAGIC || + from->magic == XFS_DIR3_FREE_MAGIC); + + if (from->magic == XFS_DIR2_FREE_MAGIC) { + to->hdr.magic = cpu_to_be32(from->magic); + to->hdr.firstdb = cpu_to_be32(from->firstdb); + to->hdr.nvalid = cpu_to_be32(from->nvalid); + to->hdr.nused = cpu_to_be32(from->nused); + } else { + struct xfs_dir3_free_hdr *hdr3 = (struct xfs_dir3_free_hdr *)to; + + hdr3->hdr.magic = cpu_to_be32(from->magic); + hdr3->firstdb = cpu_to_be32(from->firstdb); + hdr3->nvalid = cpu_to_be32(from->nvalid); + hdr3->nused = cpu_to_be32(from->nused); + } +} + +static int +xfs_dir3_free_get_buf( + struct xfs_trans *tp, + struct xfs_inode *dp, + xfs_dir2_db_t fbno, + struct xfs_buf **bpp) +{ + struct xfs_mount *mp = dp->i_mount; + struct xfs_buf *bp; + int error; + struct xfs_dir3_icfree_hdr hdr; + + error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fbno), + -1, &bp, XFS_DATA_FORK); + if (error) + return error; + + xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_FREE_BUF); + bp->b_ops = &xfs_dir3_free_buf_ops; + + /* + * Initialize the new block to be empty, and remember + * its first slot as our empty slot. + */ + memset(bp->b_addr, 0, sizeof(struct xfs_dir3_free_hdr)); + memset(&hdr, 0, sizeof(hdr)); + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + struct xfs_dir3_free_hdr *hdr3 = bp->b_addr; + + hdr.magic = XFS_DIR3_FREE_MAGIC; + + hdr3->hdr.blkno = cpu_to_be64(bp->b_bn); + hdr3->hdr.owner = cpu_to_be64(dp->i_ino); + uuid_copy(&hdr3->hdr.uuid, &mp->m_sb.sb_uuid); + } else + hdr.magic = XFS_DIR2_FREE_MAGIC; + xfs_dir3_free_hdr_to_disk(bp->b_addr, &hdr); + *bpp = bp; + return 0; +} + +/* * Log entries from a freespace block. */ STATIC void xfs_dir2_free_log_bests( - xfs_trans_t *tp, /* transaction pointer */ - xfs_dabuf_t *bp, /* freespace buffer */ + struct xfs_trans *tp, + struct xfs_buf *bp, int first, /* first entry to log */ int last) /* last entry to log */ { xfs_dir2_free_t *free; /* freespace structure */ - - free = bp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); - xfs_da_log_buf(tp, bp, - (uint)((char *)&free->bests[first] - (char *)free), - (uint)((char *)&free->bests[last] - (char *)free + - sizeof(free->bests[0]) - 1)); + __be16 *bests; + + free = bp->b_addr; + bests = xfs_dir3_free_bests_p(tp->t_mountp, free); + ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) || + free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC)); + xfs_trans_log_buf(tp, bp, + (uint)((char *)&bests[first] - (char *)free), + (uint)((char *)&bests[last] - (char *)free + + sizeof(bests[0]) - 1)); } /* @@ -80,15 +309,15 @@ xfs_dir2_free_log_bests( */ static void xfs_dir2_free_log_header( - xfs_trans_t *tp, /* transaction pointer */ - xfs_dabuf_t *bp) /* freespace buffer */ + struct xfs_trans *tp, + struct xfs_buf *bp) { xfs_dir2_free_t *free; /* freespace structure */ - free = bp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); - xfs_da_log_buf(tp, bp, (uint)((char *)&free->hdr - (char *)free), - (uint)(sizeof(xfs_dir2_free_hdr_t) - 1)); + free = bp->b_addr; + ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) || + free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC)); + xfs_trans_log_buf(tp, bp, 0, xfs_dir3_free_hdr_size(tp->t_mountp) - 1); } /* @@ -99,11 +328,11 @@ xfs_dir2_free_log_header( int /* error */ xfs_dir2_leaf_to_node( xfs_da_args_t *args, /* operation arguments */ - xfs_dabuf_t *lbp) /* leaf buffer */ + struct xfs_buf *lbp) /* leaf buffer */ { xfs_inode_t *dp; /* incore directory inode */ int error; /* error return value */ - xfs_dabuf_t *fbp; /* freespace buffer */ + struct xfs_buf *fbp; /* freespace buffer */ xfs_dir2_db_t fdb; /* freespace block number */ xfs_dir2_free_t *free; /* freespace structure */ __be16 *from; /* pointer to freespace entry */ @@ -115,6 +344,7 @@ xfs_dir2_leaf_to_node( xfs_dir2_data_off_t off; /* freespace entry value */ __be16 *to; /* pointer to freespace entry */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_dir3_icfree_hdr freehdr; trace_xfs_dir2_leaf_to_node(args); @@ -131,41 +361,53 @@ xfs_dir2_leaf_to_node( /* * Get the buffer for the new freespace block. */ - if ((error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), -1, &fbp, - XFS_DATA_FORK))) { + error = xfs_dir3_free_get_buf(tp, dp, fdb, &fbp); + if (error) return error; - } - ASSERT(fbp != NULL); - free = fbp->data; - leaf = lbp->data; + + free = fbp->b_addr; + xfs_dir3_free_hdr_from_disk(&freehdr, free); + leaf = lbp->b_addr; ltp = xfs_dir2_leaf_tail_p(mp, leaf); - /* - * Initialize the freespace block header. - */ - free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC); - free->hdr.firstdb = 0; - ASSERT(be32_to_cpu(ltp->bestcount) <= (uint)dp->i_d.di_size / mp->m_dirblksize); - free->hdr.nvalid = ltp->bestcount; + ASSERT(be32_to_cpu(ltp->bestcount) <= + (uint)dp->i_d.di_size / mp->m_dirblksize); + /* * Copy freespace entries from the leaf block to the new block. * Count active entries. */ - for (i = n = 0, from = xfs_dir2_leaf_bests_p(ltp), to = free->bests; - i < be32_to_cpu(ltp->bestcount); i++, from++, to++) { + from = xfs_dir2_leaf_bests_p(ltp); + to = xfs_dir3_free_bests_p(mp, free); + for (i = n = 0; i < be32_to_cpu(ltp->bestcount); i++, from++, to++) { if ((off = be16_to_cpu(*from)) != NULLDATAOFF) n++; *to = cpu_to_be16(off); } - free->hdr.nused = cpu_to_be32(n); - leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC); + /* - * Log everything. + * Now initialize the freespace block header. */ - xfs_dir2_leaf_log_header(tp, lbp); + freehdr.nused = n; + freehdr.nvalid = be32_to_cpu(ltp->bestcount); + + xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr); + xfs_dir2_free_log_bests(tp, fbp, 0, freehdr.nvalid - 1); xfs_dir2_free_log_header(tp, fbp); - xfs_dir2_free_log_bests(tp, fbp, 0, be32_to_cpu(free->hdr.nvalid) - 1); - xfs_da_buf_done(fbp); - xfs_dir2_leafn_check(dp, lbp); + + /* + * Converting the leaf to a leafnode is just a matter of changing the + * magic number and the ops. Do the change directly to the buffer as + * it's less work (and less code) than decoding the header to host + * format and back again. + */ + if (leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC)) + leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC); + else + leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC); + lbp->b_ops = &xfs_dir3_leafn_buf_ops; + xfs_trans_buf_set_type(tp, lbp, XFS_BLFT_DIR_LEAFN_BUF); + xfs_dir3_leaf_log_header(tp, lbp); + xfs_dir3_leaf_check(mp, lbp); return 0; } @@ -175,7 +417,7 @@ xfs_dir2_leaf_to_node( */ static int /* error */ xfs_dir2_leafn_add( - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ xfs_da_args_t *args, /* operation arguments */ int index) /* insertion pt for new entry */ { @@ -189,13 +431,17 @@ xfs_dir2_leafn_add( int lowstale; /* previous stale entry */ xfs_mount_t *mp; /* filesystem mount point */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_dir3_icleaf_hdr leafhdr; + struct xfs_dir2_leaf_entry *ents; trace_xfs_dir2_leafn_add(args, index); dp = args->dp; mp = dp->i_mount; tp = args->trans; - leaf = bp->data; + leaf = bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); /* * Quick check just to make sure we are not going to index @@ -211,15 +457,15 @@ xfs_dir2_leafn_add( * a compact. */ - if (be16_to_cpu(leaf->hdr.count) == xfs_dir2_max_leaf_ents(mp)) { - if (!leaf->hdr.stale) + if (leafhdr.count == xfs_dir3_max_leaf_ents(mp, leaf)) { + if (!leafhdr.stale) return XFS_ERROR(ENOSPC); - compact = be16_to_cpu(leaf->hdr.stale) > 1; + compact = leafhdr.stale > 1; } else compact = 0; - ASSERT(index == 0 || be32_to_cpu(leaf->ents[index - 1].hashval) <= args->hashval); - ASSERT(index == be16_to_cpu(leaf->hdr.count) || - be32_to_cpu(leaf->ents[index].hashval) >= args->hashval); + ASSERT(index == 0 || be32_to_cpu(ents[index - 1].hashval) <= args->hashval); + ASSERT(index == leafhdr.count || + be32_to_cpu(ents[index].hashval) >= args->hashval); if (args->op_flags & XFS_DA_OP_JUSTCHECK) return 0; @@ -228,61 +474,51 @@ xfs_dir2_leafn_add( * Compact out all but one stale leaf entry. Leaves behind * the entry closest to index. */ - if (compact) { - xfs_dir2_leaf_compact_x1(bp, &index, &lowstale, &highstale, - &lfloglow, &lfloghigh); - } - /* - * Set impossible logging indices for this case. - */ - else if (leaf->hdr.stale) { - lfloglow = be16_to_cpu(leaf->hdr.count); + if (compact) + xfs_dir3_leaf_compact_x1(&leafhdr, ents, &index, &lowstale, + &highstale, &lfloglow, &lfloghigh); + else if (leafhdr.stale) { + /* + * Set impossible logging indices for this case. + */ + lfloglow = leafhdr.count; lfloghigh = -1; } /* * Insert the new entry, log everything. */ - lep = xfs_dir2_leaf_find_entry(leaf, index, compact, lowstale, + lep = xfs_dir3_leaf_find_entry(&leafhdr, ents, index, compact, lowstale, highstale, &lfloglow, &lfloghigh); lep->hashval = cpu_to_be32(args->hashval); lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp, args->blkno, args->index)); - xfs_dir2_leaf_log_header(tp, bp); - xfs_dir2_leaf_log_ents(tp, bp, lfloglow, lfloghigh); - xfs_dir2_leafn_check(dp, bp); + + xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr); + xfs_dir3_leaf_log_header(tp, bp); + xfs_dir3_leaf_log_ents(tp, bp, lfloglow, lfloghigh); + xfs_dir3_leaf_check(mp, bp); return 0; } #ifdef DEBUG -/* - * Check internal consistency of a leafn block. - */ -void -xfs_dir2_leafn_check( - xfs_inode_t *dp, /* incore directory inode */ - xfs_dabuf_t *bp) /* leaf buffer */ +static void +xfs_dir2_free_hdr_check( + struct xfs_mount *mp, + struct xfs_buf *bp, + xfs_dir2_db_t db) { - int i; /* leaf index */ - xfs_dir2_leaf_t *leaf; /* leaf structure */ - xfs_mount_t *mp; /* filesystem mount point */ - int stale; /* count of stale leaves */ + struct xfs_dir3_icfree_hdr hdr; - leaf = bp->data; - mp = dp->i_mount; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - ASSERT(be16_to_cpu(leaf->hdr.count) <= xfs_dir2_max_leaf_ents(mp)); - for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) { - if (i + 1 < be16_to_cpu(leaf->hdr.count)) { - ASSERT(be32_to_cpu(leaf->ents[i].hashval) <= - be32_to_cpu(leaf->ents[i + 1].hashval)); - } - if (leaf->ents[i].address == cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) - stale++; - } - ASSERT(be16_to_cpu(leaf->hdr.stale) == stale); + xfs_dir3_free_hdr_from_disk(&hdr, bp->b_addr); + + ASSERT((hdr.firstdb % xfs_dir3_free_max_bests(mp)) == 0); + ASSERT(hdr.firstdb <= db); + ASSERT(db < hdr.firstdb + hdr.nvalid); } +#else +#define xfs_dir2_free_hdr_check(mp, dp, db) #endif /* DEBUG */ /* @@ -291,18 +527,25 @@ xfs_dir2_leafn_check( */ xfs_dahash_t /* hash value */ xfs_dir2_leafn_lasthash( - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ int *count) /* count of entries in leaf */ { - xfs_dir2_leaf_t *leaf; /* leaf structure */ + struct xfs_dir2_leaf *leaf = bp->b_addr; + struct xfs_dir2_leaf_entry *ents; + struct xfs_dir3_icleaf_hdr leafhdr; + + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + + ASSERT(leafhdr.magic == XFS_DIR2_LEAFN_MAGIC || + leafhdr.magic == XFS_DIR3_LEAFN_MAGIC); - leaf = bp->data; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); if (count) - *count = be16_to_cpu(leaf->hdr.count); - if (!leaf->hdr.count) + *count = leafhdr.count; + if (!leafhdr.count) return 0; - return be32_to_cpu(leaf->ents[be16_to_cpu(leaf->hdr.count) - 1].hashval); + + ents = xfs_dir3_leaf_ents_p(leaf); + return be32_to_cpu(ents[leafhdr.count - 1].hashval); } /* @@ -311,12 +554,12 @@ xfs_dir2_leafn_lasthash( */ STATIC int xfs_dir2_leafn_lookup_for_addname( - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ xfs_da_args_t *args, /* operation arguments */ int *indexp, /* out: leaf entry index */ xfs_da_state_t *state) /* state to fill in */ { - xfs_dabuf_t *curbp = NULL; /* current data/free buffer */ + struct xfs_buf *curbp = NULL; /* current data/free buffer */ xfs_dir2_db_t curdb = -1; /* current data block number */ xfs_dir2_db_t curfdb = -1; /* current free block number */ xfs_inode_t *dp; /* incore directory inode */ @@ -331,16 +574,19 @@ xfs_dir2_leafn_lookup_for_addname( xfs_dir2_db_t newdb; /* new data block number */ xfs_dir2_db_t newfdb; /* new free block number */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_dir2_leaf_entry *ents; + struct xfs_dir3_icleaf_hdr leafhdr; dp = args->dp; tp = args->trans; mp = dp->i_mount; - leaf = bp->data; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); -#ifdef __KERNEL__ - ASSERT(be16_to_cpu(leaf->hdr.count) > 0); -#endif - xfs_dir2_leafn_check(dp, bp); + leaf = bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); + + xfs_dir3_leaf_check(mp, bp); + ASSERT(leafhdr.count > 0); + /* * Look up the hash value in the leaf entries. */ @@ -352,16 +598,17 @@ xfs_dir2_leafn_lookup_for_addname( /* If so, it's a free block buffer, get the block number. */ curbp = state->extrablk.bp; curfdb = state->extrablk.blkno; - free = curbp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); + free = curbp->b_addr; + ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) || + free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC)); } length = xfs_dir2_data_entsize(args->namelen); /* * Loop over leaf entries with the right hash value. */ - for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) && - be32_to_cpu(lep->hashval) == args->hashval; - lep++, index++) { + for (lep = &ents[index]; + index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval; + lep++, index++) { /* * Skip stale leaf entries. */ @@ -380,6 +627,8 @@ xfs_dir2_leafn_lookup_for_addname( * in hand, take a look at it. */ if (newdb != curdb) { + __be16 *bests; + curdb = newdb; /* * Convert the data block to the free block @@ -394,23 +643,16 @@ xfs_dir2_leafn_lookup_for_addname( * If we had one before, drop it. */ if (curbp) - xfs_da_brelse(tp, curbp); - /* - * Read the free block. - */ - error = xfs_da_read_buf(tp, dp, + xfs_trans_brelse(tp, curbp); + + error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(mp, newfdb), - -1, &curbp, XFS_DATA_FORK); + &curbp); if (error) return error; - free = curbp->data; - ASSERT(be32_to_cpu(free->hdr.magic) == - XFS_DIR2_FREE_MAGIC); - ASSERT((be32_to_cpu(free->hdr.firstdb) % - xfs_dir2_free_max_bests(mp)) == 0); - ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb); - ASSERT(curdb < be32_to_cpu(free->hdr.firstdb) + - be32_to_cpu(free->hdr.nvalid)); + free = curbp->b_addr; + + xfs_dir2_free_hdr_check(mp, curbp, curdb); } /* * Get the index for our entry. @@ -419,16 +661,16 @@ xfs_dir2_leafn_lookup_for_addname( /* * If it has room, return it. */ - if (unlikely(free->bests[fi] == - cpu_to_be16(NULLDATAOFF))) { + bests = xfs_dir3_free_bests_p(mp, free); + if (unlikely(bests[fi] == cpu_to_be16(NULLDATAOFF))) { XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int", XFS_ERRLEVEL_LOW, mp); if (curfdb != newfdb) - xfs_da_brelse(tp, curbp); + xfs_trans_brelse(tp, curbp); return XFS_ERROR(EFSCORRUPTED); } curfdb = newfdb; - if (be16_to_cpu(free->bests[fi]) >= length) + if (be16_to_cpu(bests[fi]) >= length) goto out; } } @@ -442,6 +684,12 @@ out: state->extrablk.bp = curbp; state->extrablk.index = fi; state->extrablk.blkno = curfdb; + + /* + * Important: this magic number is not in the buffer - it's for + * buffer type information and therefore only the free/data type + * matters here, not whether CRCs are enabled or not. + */ state->extrablk.magic = XFS_DIR2_FREE_MAGIC; } else { state->extravalid = 0; @@ -459,12 +707,12 @@ out: */ STATIC int xfs_dir2_leafn_lookup_for_entry( - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ xfs_da_args_t *args, /* operation arguments */ int *indexp, /* out: leaf entry index */ xfs_da_state_t *state) /* state to fill in */ { - xfs_dabuf_t *curbp = NULL; /* current data/free buffer */ + struct xfs_buf *curbp = NULL; /* current data/free buffer */ xfs_dir2_db_t curdb = -1; /* current data block number */ xfs_dir2_data_entry_t *dep; /* data block entry */ xfs_inode_t *dp; /* incore directory inode */ @@ -476,16 +724,19 @@ xfs_dir2_leafn_lookup_for_entry( xfs_dir2_db_t newdb; /* new data block number */ xfs_trans_t *tp; /* transaction pointer */ enum xfs_dacmp cmp; /* comparison result */ + struct xfs_dir2_leaf_entry *ents; + struct xfs_dir3_icleaf_hdr leafhdr; dp = args->dp; tp = args->trans; mp = dp->i_mount; - leaf = bp->data; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); -#ifdef __KERNEL__ - ASSERT(be16_to_cpu(leaf->hdr.count) > 0); -#endif - xfs_dir2_leafn_check(dp, bp); + leaf = bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); + + xfs_dir3_leaf_check(mp, bp); + ASSERT(leafhdr.count > 0); + /* * Look up the hash value in the leaf entries. */ @@ -500,9 +751,9 @@ xfs_dir2_leafn_lookup_for_entry( /* * Loop over leaf entries with the right hash value. */ - for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) && - be32_to_cpu(lep->hashval) == args->hashval; - lep++, index++) { + for (lep = &ents[index]; + index < leafhdr.count && be32_to_cpu(lep->hashval) == args->hashval; + lep++, index++) { /* * Skip stale leaf entries. */ @@ -525,7 +776,7 @@ xfs_dir2_leafn_lookup_for_entry( */ if (curbp && (args->cmpresult == XFS_CMP_DIFFERENT || curdb != state->extrablk.blkno)) - xfs_da_brelse(tp, curbp); + xfs_trans_brelse(tp, curbp); /* * If needing the block that is saved with a CI match, * use it otherwise read in the new data block. @@ -535,19 +786,19 @@ xfs_dir2_leafn_lookup_for_entry( ASSERT(state->extravalid); curbp = state->extrablk.bp; } else { - error = xfs_da_read_buf(tp, dp, + error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, newdb), - -1, &curbp, XFS_DATA_FORK); + -1, &curbp); if (error) return error; } - xfs_dir2_data_check(dp, curbp); + xfs_dir3_data_check(dp, curbp); curdb = newdb; } /* * Point to the data entry. */ - dep = (xfs_dir2_data_entry_t *)((char *)curbp->data + + dep = (xfs_dir2_data_entry_t *)((char *)curbp->b_addr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address))); /* * Compare the entry and if it's an exact match, return @@ -559,7 +810,7 @@ xfs_dir2_leafn_lookup_for_entry( /* If there is a CI match block, drop it */ if (args->cmpresult != XFS_CMP_DIFFERENT && curdb != state->extrablk.blkno) - xfs_da_brelse(tp, state->extrablk.bp); + xfs_trans_brelse(tp, state->extrablk.bp); args->cmpresult = cmp; args->inumber = be64_to_cpu(dep->inumber); *indexp = index; @@ -567,14 +818,15 @@ xfs_dir2_leafn_lookup_for_entry( state->extrablk.bp = curbp; state->extrablk.blkno = curdb; state->extrablk.index = (int)((char *)dep - - (char *)curbp->data); + (char *)curbp->b_addr); state->extrablk.magic = XFS_DIR2_DATA_MAGIC; + curbp->b_ops = &xfs_dir3_data_buf_ops; + xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF); if (cmp == XFS_CMP_EXACT) return XFS_ERROR(EEXIST); } } - ASSERT(index == be16_to_cpu(leaf->hdr.count) || - (args->op_flags & XFS_DA_OP_OKNOENT)); + ASSERT(index == leafhdr.count || (args->op_flags & XFS_DA_OP_OKNOENT)); if (curbp) { if (args->cmpresult == XFS_CMP_DIFFERENT) { /* Giving back last used data block. */ @@ -583,10 +835,12 @@ xfs_dir2_leafn_lookup_for_entry( state->extrablk.index = -1; state->extrablk.blkno = curdb; state->extrablk.magic = XFS_DIR2_DATA_MAGIC; + curbp->b_ops = &xfs_dir3_data_buf_ops; + xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF); } else { /* If the curbp is not the CI match block, drop it */ if (state->extrablk.bp != curbp) - xfs_da_brelse(tp, curbp); + xfs_trans_brelse(tp, curbp); } } else { state->extravalid = 0; @@ -602,7 +856,7 @@ xfs_dir2_leafn_lookup_for_entry( */ int xfs_dir2_leafn_lookup_int( - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ xfs_da_args_t *args, /* operation arguments */ int *indexp, /* out: leaf entry index */ xfs_da_state_t *state) /* state to fill in */ @@ -618,52 +872,50 @@ xfs_dir2_leafn_lookup_int( * Log entries and headers. Stale entries are preserved. */ static void -xfs_dir2_leafn_moveents( - xfs_da_args_t *args, /* operation arguments */ - xfs_dabuf_t *bp_s, /* source leaf buffer */ - int start_s, /* source leaf index */ - xfs_dabuf_t *bp_d, /* destination leaf buffer */ - int start_d, /* destination leaf index */ - int count) /* count of leaves to copy */ +xfs_dir3_leafn_moveents( + xfs_da_args_t *args, /* operation arguments */ + struct xfs_buf *bp_s, /* source */ + struct xfs_dir3_icleaf_hdr *shdr, + struct xfs_dir2_leaf_entry *sents, + int start_s,/* source leaf index */ + struct xfs_buf *bp_d, /* destination */ + struct xfs_dir3_icleaf_hdr *dhdr, + struct xfs_dir2_leaf_entry *dents, + int start_d,/* destination leaf index */ + int count) /* count of leaves to copy */ { - xfs_dir2_leaf_t *leaf_d; /* destination leaf structure */ - xfs_dir2_leaf_t *leaf_s; /* source leaf structure */ - int stale; /* count stale leaves copied */ - xfs_trans_t *tp; /* transaction pointer */ + struct xfs_trans *tp = args->trans; + int stale; /* count stale leaves copied */ trace_xfs_dir2_leafn_moveents(args, start_s, start_d, count); /* * Silently return if nothing to do. */ - if (count == 0) { + if (count == 0) return; - } - tp = args->trans; - leaf_s = bp_s->data; - leaf_d = bp_d->data; + /* * If the destination index is not the end of the current * destination leaf entries, open up a hole in the destination * to hold the new entries. */ - if (start_d < be16_to_cpu(leaf_d->hdr.count)) { - memmove(&leaf_d->ents[start_d + count], &leaf_d->ents[start_d], - (be16_to_cpu(leaf_d->hdr.count) - start_d) * - sizeof(xfs_dir2_leaf_entry_t)); - xfs_dir2_leaf_log_ents(tp, bp_d, start_d + count, - count + be16_to_cpu(leaf_d->hdr.count) - 1); + if (start_d < dhdr->count) { + memmove(&dents[start_d + count], &dents[start_d], + (dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t)); + xfs_dir3_leaf_log_ents(tp, bp_d, start_d + count, + count + dhdr->count - 1); } /* * If the source has stale leaves, count the ones in the copy range * so we can update the header correctly. */ - if (leaf_s->hdr.stale) { + if (shdr->stale) { int i; /* temp leaf index */ for (i = start_s, stale = 0; i < start_s + count; i++) { - if (leaf_s->ents[i].address == - cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) + if (sents[i].address == + cpu_to_be32(XFS_DIR2_NULL_DATAPTR)) stale++; } } else @@ -671,29 +923,27 @@ xfs_dir2_leafn_moveents( /* * Copy the leaf entries from source to destination. */ - memcpy(&leaf_d->ents[start_d], &leaf_s->ents[start_s], + memcpy(&dents[start_d], &sents[start_s], count * sizeof(xfs_dir2_leaf_entry_t)); - xfs_dir2_leaf_log_ents(tp, bp_d, start_d, start_d + count - 1); + xfs_dir3_leaf_log_ents(tp, bp_d, start_d, start_d + count - 1); + /* * If there are source entries after the ones we copied, * delete the ones we copied by sliding the next ones down. */ - if (start_s + count < be16_to_cpu(leaf_s->hdr.count)) { - memmove(&leaf_s->ents[start_s], &leaf_s->ents[start_s + count], + if (start_s + count < shdr->count) { + memmove(&sents[start_s], &sents[start_s + count], count * sizeof(xfs_dir2_leaf_entry_t)); - xfs_dir2_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1); + xfs_dir3_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1); } + /* * Update the headers and log them. */ - be16_add_cpu(&leaf_s->hdr.count, -(count)); - be16_add_cpu(&leaf_s->hdr.stale, -(stale)); - be16_add_cpu(&leaf_d->hdr.count, count); - be16_add_cpu(&leaf_d->hdr.stale, stale); - xfs_dir2_leaf_log_header(tp, bp_s); - xfs_dir2_leaf_log_header(tp, bp_d); - xfs_dir2_leafn_check(args->dp, bp_s); - xfs_dir2_leafn_check(args->dp, bp_d); + shdr->count -= count; + shdr->stale -= stale; + dhdr->count += count; + dhdr->stale += stale; } /* @@ -702,21 +952,25 @@ xfs_dir2_leafn_moveents( */ int /* sort order */ xfs_dir2_leafn_order( - xfs_dabuf_t *leaf1_bp, /* leaf1 buffer */ - xfs_dabuf_t *leaf2_bp) /* leaf2 buffer */ + struct xfs_buf *leaf1_bp, /* leaf1 buffer */ + struct xfs_buf *leaf2_bp) /* leaf2 buffer */ { - xfs_dir2_leaf_t *leaf1; /* leaf1 structure */ - xfs_dir2_leaf_t *leaf2; /* leaf2 structure */ - - leaf1 = leaf1_bp->data; - leaf2 = leaf2_bp->data; - ASSERT(leaf1->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - if (be16_to_cpu(leaf1->hdr.count) > 0 && - be16_to_cpu(leaf2->hdr.count) > 0 && - (be32_to_cpu(leaf2->ents[0].hashval) < be32_to_cpu(leaf1->ents[0].hashval) || - be32_to_cpu(leaf2->ents[be16_to_cpu(leaf2->hdr.count) - 1].hashval) < - be32_to_cpu(leaf1->ents[be16_to_cpu(leaf1->hdr.count) - 1].hashval))) + struct xfs_dir2_leaf *leaf1 = leaf1_bp->b_addr; + struct xfs_dir2_leaf *leaf2 = leaf2_bp->b_addr; + struct xfs_dir2_leaf_entry *ents1; + struct xfs_dir2_leaf_entry *ents2; + struct xfs_dir3_icleaf_hdr hdr1; + struct xfs_dir3_icleaf_hdr hdr2; + + xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1); + xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2); + ents1 = xfs_dir3_leaf_ents_p(leaf1); + ents2 = xfs_dir3_leaf_ents_p(leaf2); + + if (hdr1.count > 0 && hdr2.count > 0 && + (be32_to_cpu(ents2[0].hashval) < be32_to_cpu(ents1[0].hashval) || + be32_to_cpu(ents2[hdr2.count - 1].hashval) < + be32_to_cpu(ents1[hdr1.count - 1].hashval))) return 1; return 0; } @@ -740,11 +994,15 @@ xfs_dir2_leafn_rebalance( xfs_dir2_leaf_t *leaf1; /* first leaf structure */ xfs_dir2_leaf_t *leaf2; /* second leaf structure */ int mid; /* midpoint leaf index */ -#ifdef DEBUG +#if defined(DEBUG) || defined(XFS_WARN) int oldstale; /* old count of stale leaves */ #endif int oldsum; /* old total leaf count */ int swap; /* swapped leaf blocks */ + struct xfs_dir2_leaf_entry *ents1; + struct xfs_dir2_leaf_entry *ents2; + struct xfs_dir3_icleaf_hdr hdr1; + struct xfs_dir3_icleaf_hdr hdr2; args = state->args; /* @@ -757,13 +1015,19 @@ xfs_dir2_leafn_rebalance( blk1 = blk2; blk2 = tmp; } - leaf1 = blk1->bp->data; - leaf2 = blk2->bp->data; - oldsum = be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count); -#ifdef DEBUG - oldstale = be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale); + leaf1 = blk1->bp->b_addr; + leaf2 = blk2->bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&hdr1, leaf1); + xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf2); + ents1 = xfs_dir3_leaf_ents_p(leaf1); + ents2 = xfs_dir3_leaf_ents_p(leaf2); + + oldsum = hdr1.count + hdr2.count; +#if defined(DEBUG) || defined(XFS_WARN) + oldstale = hdr1.stale + hdr2.stale; #endif mid = oldsum >> 1; + /* * If the old leaf count was odd then the new one will be even, * so we need to divide the new count evenly. @@ -771,10 +1035,10 @@ xfs_dir2_leafn_rebalance( if (oldsum & 1) { xfs_dahash_t midhash; /* middle entry hash value */ - if (mid >= be16_to_cpu(leaf1->hdr.count)) - midhash = be32_to_cpu(leaf2->ents[mid - be16_to_cpu(leaf1->hdr.count)].hashval); + if (mid >= hdr1.count) + midhash = be32_to_cpu(ents2[mid - hdr1.count].hashval); else - midhash = be32_to_cpu(leaf1->ents[mid].hashval); + midhash = be32_to_cpu(ents1[mid].hashval); isleft = args->hashval <= midhash; } /* @@ -788,30 +1052,42 @@ xfs_dir2_leafn_rebalance( * Calculate moved entry count. Positive means left-to-right, * negative means right-to-left. Then move the entries. */ - count = be16_to_cpu(leaf1->hdr.count) - mid + (isleft == 0); + count = hdr1.count - mid + (isleft == 0); if (count > 0) - xfs_dir2_leafn_moveents(args, blk1->bp, - be16_to_cpu(leaf1->hdr.count) - count, blk2->bp, 0, count); + xfs_dir3_leafn_moveents(args, blk1->bp, &hdr1, ents1, + hdr1.count - count, blk2->bp, + &hdr2, ents2, 0, count); else if (count < 0) - xfs_dir2_leafn_moveents(args, blk2->bp, 0, blk1->bp, - be16_to_cpu(leaf1->hdr.count), count); - ASSERT(be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count) == oldsum); - ASSERT(be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale) == oldstale); + xfs_dir3_leafn_moveents(args, blk2->bp, &hdr2, ents2, 0, + blk1->bp, &hdr1, ents1, + hdr1.count, count); + + ASSERT(hdr1.count + hdr2.count == oldsum); + ASSERT(hdr1.stale + hdr2.stale == oldstale); + + /* log the changes made when moving the entries */ + xfs_dir3_leaf_hdr_to_disk(leaf1, &hdr1); + xfs_dir3_leaf_hdr_to_disk(leaf2, &hdr2); + xfs_dir3_leaf_log_header(args->trans, blk1->bp); + xfs_dir3_leaf_log_header(args->trans, blk2->bp); + + xfs_dir3_leaf_check(args->dp->i_mount, blk1->bp); + xfs_dir3_leaf_check(args->dp->i_mount, blk2->bp); + /* * Mark whether we're inserting into the old or new leaf. */ - if (be16_to_cpu(leaf1->hdr.count) < be16_to_cpu(leaf2->hdr.count)) + if (hdr1.count < hdr2.count) state->inleaf = swap; - else if (be16_to_cpu(leaf1->hdr.count) > be16_to_cpu(leaf2->hdr.count)) + else if (hdr1.count > hdr2.count) state->inleaf = !swap; else - state->inleaf = - swap ^ (blk1->index <= be16_to_cpu(leaf1->hdr.count)); + state->inleaf = swap ^ (blk1->index <= hdr1.count); /* * Adjust the expected index for insertion. */ if (!state->inleaf) - blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count); + blk2->index = blk1->index - hdr1.count; /* * Finally sanity check just to make sure we are not returning a @@ -826,6 +1102,86 @@ xfs_dir2_leafn_rebalance( } } +static int +xfs_dir3_data_block_free( + xfs_da_args_t *args, + struct xfs_dir2_data_hdr *hdr, + struct xfs_dir2_free *free, + xfs_dir2_db_t fdb, + int findex, + struct xfs_buf *fbp, + int longest) +{ + struct xfs_trans *tp = args->trans; + int logfree = 0; + __be16 *bests; + struct xfs_dir3_icfree_hdr freehdr; + + xfs_dir3_free_hdr_from_disk(&freehdr, free); + + bests = xfs_dir3_free_bests_p(tp->t_mountp, free); + if (hdr) { + /* + * Data block is not empty, just set the free entry to the new + * value. + */ + bests[findex] = cpu_to_be16(longest); + xfs_dir2_free_log_bests(tp, fbp, findex, findex); + return 0; + } + + /* One less used entry in the free table. */ + freehdr.nused--; + + /* + * If this was the last entry in the table, we can trim the table size + * back. There might be other entries at the end referring to + * non-existent data blocks, get those too. + */ + if (findex == freehdr.nvalid - 1) { + int i; /* free entry index */ + + for (i = findex - 1; i >= 0; i--) { + if (bests[i] != cpu_to_be16(NULLDATAOFF)) + break; + } + freehdr.nvalid = i + 1; + logfree = 0; + } else { + /* Not the last entry, just punch it out. */ + bests[findex] = cpu_to_be16(NULLDATAOFF); + logfree = 1; + } + + xfs_dir3_free_hdr_to_disk(free, &freehdr); + xfs_dir2_free_log_header(tp, fbp); + + /* + * If there are no useful entries left in the block, get rid of the + * block if we can. + */ + if (!freehdr.nused) { + int error; + + error = xfs_dir2_shrink_inode(args, fdb, fbp); + if (error == 0) { + fbp = NULL; + logfree = 0; + } else if (error != ENOSPC || args->total != 0) + return error; + /* + * It's possible to get ENOSPC if there is no + * space reservation. In this case some one + * else will eventually get rid of this block. + */ + } + + /* Log the free entry that changed, unless we got rid of it. */ + if (logfree) + xfs_dir2_free_log_bests(tp, fbp, findex, findex); + return 0; +} + /* * Remove an entry from a node directory. * This removes the leaf entry and the data entry, @@ -834,14 +1190,14 @@ xfs_dir2_leafn_rebalance( static int /* error */ xfs_dir2_leafn_remove( xfs_da_args_t *args, /* operation arguments */ - xfs_dabuf_t *bp, /* leaf buffer */ + struct xfs_buf *bp, /* leaf buffer */ int index, /* leaf entry index */ xfs_da_state_blk_t *dblk, /* data block */ int *rval) /* resulting block needs join */ { xfs_dir2_data_hdr_t *hdr; /* data block header */ xfs_dir2_db_t db; /* data block number */ - xfs_dabuf_t *dbp; /* data block buffer */ + struct xfs_buf *dbp; /* data block buffer */ xfs_dir2_data_entry_t *dep; /* data block entry */ xfs_inode_t *dp; /* incore directory inode */ xfs_dir2_leaf_t *leaf; /* leaf structure */ @@ -852,18 +1208,24 @@ xfs_dir2_leafn_remove( int needlog; /* need to log data header */ int needscan; /* need to rescan data frees */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_dir2_data_free *bf; /* bestfree table */ + struct xfs_dir3_icleaf_hdr leafhdr; + struct xfs_dir2_leaf_entry *ents; trace_xfs_dir2_leafn_remove(args, index); dp = args->dp; tp = args->trans; mp = dp->i_mount; - leaf = bp->data; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); + leaf = bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); + /* * Point to the entry we're removing. */ - lep = &leaf->ents[index]; + lep = &ents[index]; + /* * Extract the data block and offset from the entry. */ @@ -871,22 +1233,27 @@ xfs_dir2_leafn_remove( ASSERT(dblk->blkno == db); off = xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)); ASSERT(dblk->index == off); + /* * Kill the leaf entry by marking it stale. * Log the leaf block changes. */ - be16_add_cpu(&leaf->hdr.stale, 1); - xfs_dir2_leaf_log_header(tp, bp); + leafhdr.stale++; + xfs_dir3_leaf_hdr_to_disk(leaf, &leafhdr); + xfs_dir3_leaf_log_header(tp, bp); + lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR); - xfs_dir2_leaf_log_ents(tp, bp, index, index); + xfs_dir3_leaf_log_ents(tp, bp, index, index); + /* * Make the data entry free. Keep track of the longest freespace * in the data block in case it changes. */ dbp = dblk->bp; - hdr = dbp->data; + hdr = dbp->b_addr; dep = (xfs_dir2_data_entry_t *)((char *)hdr + off); - longest = be16_to_cpu(hdr->bestfree[0].length); + bf = xfs_dir3_data_bestfree_p(hdr); + longest = be16_to_cpu(bf[0].length); needlog = needscan = 0; xfs_dir2_data_make_free(tp, dbp, off, xfs_dir2_data_entsize(dep->namelen), &needlog, &needscan); @@ -898,43 +1265,47 @@ xfs_dir2_leafn_remove( xfs_dir2_data_freescan(mp, hdr, &needlog); if (needlog) xfs_dir2_data_log_header(tp, dbp); - xfs_dir2_data_check(dp, dbp); + xfs_dir3_data_check(dp, dbp); /* * If the longest data block freespace changes, need to update * the corresponding freeblock entry. */ - if (longest < be16_to_cpu(hdr->bestfree[0].length)) { + if (longest < be16_to_cpu(bf[0].length)) { int error; /* error return value */ - xfs_dabuf_t *fbp; /* freeblock buffer */ + struct xfs_buf *fbp; /* freeblock buffer */ xfs_dir2_db_t fdb; /* freeblock block number */ int findex; /* index in freeblock entries */ xfs_dir2_free_t *free; /* freeblock structure */ - int logfree; /* need to log free entry */ /* * Convert the data block number to a free block, * read in the free block. */ fdb = xfs_dir2_db_to_fdb(mp, db); - if ((error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, fdb), - -1, &fbp, XFS_DATA_FORK))) { + error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(mp, fdb), + &fbp); + if (error) return error; - } - free = fbp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); - ASSERT(be32_to_cpu(free->hdr.firstdb) == - xfs_dir2_free_max_bests(mp) * - (fdb - XFS_DIR2_FREE_FIRSTDB(mp))); + free = fbp->b_addr; +#ifdef DEBUG + { + struct xfs_dir3_icfree_hdr freehdr; + xfs_dir3_free_hdr_from_disk(&freehdr, free); + ASSERT(freehdr.firstdb == xfs_dir3_free_max_bests(mp) * + (fdb - XFS_DIR2_FREE_FIRSTDB(mp))); + } +#endif /* * Calculate which entry we need to fix. */ findex = xfs_dir2_db_to_fdindex(mp, db); - longest = be16_to_cpu(hdr->bestfree[0].length); + longest = be16_to_cpu(bf[0].length); /* * If the data block is now empty we can get rid of it * (usually). */ - if (longest == mp->m_dirblksize - (uint)sizeof(*hdr)) { + if (longest == mp->m_dirblksize - + xfs_dir3_data_entry_offset(hdr)) { /* * Try to punch out the data block. */ @@ -948,91 +1319,26 @@ xfs_dir2_leafn_remove( * In this case just drop the buffer and some one else * will eventually get rid of the empty block. */ - else if (error == ENOSPC && args->total == 0) - xfs_da_buf_done(dbp); - else + else if (!(error == ENOSPC && args->total == 0)) return error; } /* * If we got rid of the data block, we can eliminate that entry * in the free block. */ - if (hdr == NULL) { - /* - * One less used entry in the free table. - */ - be32_add_cpu(&free->hdr.nused, -1); - xfs_dir2_free_log_header(tp, fbp); - /* - * If this was the last entry in the table, we can - * trim the table size back. There might be other - * entries at the end referring to non-existent - * data blocks, get those too. - */ - if (findex == be32_to_cpu(free->hdr.nvalid) - 1) { - int i; /* free entry index */ - - for (i = findex - 1; - i >= 0 && - free->bests[i] == cpu_to_be16(NULLDATAOFF); - i--) - continue; - free->hdr.nvalid = cpu_to_be32(i + 1); - logfree = 0; - } - /* - * Not the last entry, just punch it out. - */ - else { - free->bests[findex] = cpu_to_be16(NULLDATAOFF); - logfree = 1; - } - /* - * If there are no useful entries left in the block, - * get rid of the block if we can. - */ - if (!free->hdr.nused) { - error = xfs_dir2_shrink_inode(args, fdb, fbp); - if (error == 0) { - fbp = NULL; - logfree = 0; - } else if (error != ENOSPC || args->total != 0) - return error; - /* - * It's possible to get ENOSPC if there is no - * space reservation. In this case some one - * else will eventually get rid of this block. - */ - } - } - /* - * Data block is not empty, just set the free entry to - * the new value. - */ - else { - free->bests[findex] = cpu_to_be16(longest); - logfree = 1; - } - /* - * Log the free entry that changed, unless we got rid of it. - */ - if (logfree) - xfs_dir2_free_log_bests(tp, fbp, findex, findex); - /* - * Drop the buffer if we still have it. - */ - if (fbp) - xfs_da_buf_done(fbp); + error = xfs_dir3_data_block_free(args, hdr, free, + fdb, findex, fbp, longest); + if (error) + return error; } - xfs_dir2_leafn_check(dp, bp); + + xfs_dir3_leaf_check(mp, bp); /* * Return indication of whether this leaf block is empty enough * to justify trying to join it with a neighbor. */ - *rval = - ((uint)sizeof(leaf->hdr) + - (uint)sizeof(leaf->ents[0]) * - (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale))) < + *rval = (xfs_dir3_leaf_hdr_size(leaf) + + (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) < mp->m_dir_magicpct; return 0; } @@ -1065,11 +1371,11 @@ xfs_dir2_leafn_split( /* * Initialize the new leaf block. */ - error = xfs_dir2_leaf_init(args, xfs_dir2_da_to_db(mp, blkno), - &newblk->bp, XFS_DIR2_LEAFN_MAGIC); - if (error) { + error = xfs_dir3_leaf_get_buf(args, xfs_dir2_da_to_db(mp, blkno), + &newblk->bp, XFS_DIR2_LEAFN_MAGIC); + if (error) return error; - } + newblk->blkno = blkno; newblk->magic = XFS_DIR2_LEAFN_MAGIC; /* @@ -1077,7 +1383,7 @@ xfs_dir2_leafn_split( * block into the leaves. */ xfs_dir2_leafn_rebalance(state, oldblk, newblk); - error = xfs_da_blk_link(state, oldblk, newblk); + error = xfs_da3_blk_link(state, oldblk, newblk); if (error) { return error; } @@ -1093,8 +1399,8 @@ xfs_dir2_leafn_split( */ oldblk->hashval = xfs_dir2_leafn_lasthash(oldblk->bp, NULL); newblk->hashval = xfs_dir2_leafn_lasthash(newblk->bp, NULL); - xfs_dir2_leafn_check(args->dp, oldblk->bp); - xfs_dir2_leafn_check(args->dp, newblk->bp); + xfs_dir3_leaf_check(mp, oldblk->bp); + xfs_dir3_leaf_check(mp, newblk->bp); return error; } @@ -1114,15 +1420,16 @@ xfs_dir2_leafn_toosmall( { xfs_da_state_blk_t *blk; /* leaf block */ xfs_dablk_t blkno; /* leaf block number */ - xfs_dabuf_t *bp; /* leaf buffer */ + struct xfs_buf *bp; /* leaf buffer */ int bytes; /* bytes in use */ int count; /* leaf live entry count */ int error; /* error return value */ int forward; /* sibling block direction */ int i; /* sibling counter */ - xfs_da_blkinfo_t *info; /* leaf block header */ xfs_dir2_leaf_t *leaf; /* leaf structure */ int rval; /* result from path_shift */ + struct xfs_dir3_icleaf_hdr leafhdr; + struct xfs_dir2_leaf_entry *ents; /* * Check for the degenerate case of the block being over 50% full. @@ -1130,11 +1437,13 @@ xfs_dir2_leafn_toosmall( * to coalesce with a sibling. */ blk = &state->path.blk[state->path.active - 1]; - info = blk->bp->data; - ASSERT(info->magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - leaf = (xfs_dir2_leaf_t *)info; - count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale); - bytes = (uint)sizeof(leaf->hdr) + count * (uint)sizeof(leaf->ents[0]); + leaf = blk->bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&leafhdr, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); + xfs_dir3_leaf_check(state->args->dp->i_mount, blk->bp); + + count = leafhdr.count - leafhdr.stale; + bytes = xfs_dir3_leaf_hdr_size(leaf) + count * sizeof(ents[0]); if (bytes > (state->blocksize >> 1)) { /* * Blk over 50%, don't try to join. @@ -1153,9 +1462,9 @@ xfs_dir2_leafn_toosmall( * Make altpath point to the block we want to keep and * path point to the block we want to drop (this one). */ - forward = (info->forw != 0); + forward = (leafhdr.forw != 0); memcpy(&state->altpath, &state->path, sizeof(state->path)); - error = xfs_da_path_shift(state, &state->altpath, forward, 0, + error = xfs_da3_path_shift(state, &state->altpath, forward, 0, &rval); if (error) return error; @@ -1169,36 +1478,39 @@ xfs_dir2_leafn_toosmall( * We prefer coalescing with the lower numbered sibling so as * to shrink a directory over time. */ - forward = be32_to_cpu(info->forw) < be32_to_cpu(info->back); + forward = leafhdr.forw < leafhdr.back; for (i = 0, bp = NULL; i < 2; forward = !forward, i++) { - blkno = forward ? be32_to_cpu(info->forw) : be32_to_cpu(info->back); + struct xfs_dir3_icleaf_hdr hdr2; + + blkno = forward ? leafhdr.forw : leafhdr.back; if (blkno == 0) continue; /* * Read the sibling leaf block. */ - if ((error = - xfs_da_read_buf(state->args->trans, state->args->dp, blkno, - -1, &bp, XFS_DATA_FORK))) { + error = xfs_dir3_leafn_read(state->args->trans, state->args->dp, + blkno, -1, &bp); + if (error) return error; - } - ASSERT(bp != NULL); + /* * Count bytes in the two blocks combined. */ - leaf = (xfs_dir2_leaf_t *)info; - count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale); + count = leafhdr.count - leafhdr.stale; bytes = state->blocksize - (state->blocksize >> 2); - leaf = bp->data; - ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - count += be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale); - bytes -= count * (uint)sizeof(leaf->ents[0]); + + leaf = bp->b_addr; + xfs_dir3_leaf_hdr_from_disk(&hdr2, leaf); + ents = xfs_dir3_leaf_ents_p(leaf); + count += hdr2.count - hdr2.stale; + bytes -= count * sizeof(ents[0]); + /* * Fits with at least 25% to spare. */ if (bytes >= 0) break; - xfs_da_brelse(state->args->trans, bp); + xfs_trans_brelse(state->args->trans, bp); } /* * Didn't like either block, give up. @@ -1207,21 +1519,17 @@ xfs_dir2_leafn_toosmall( *action = 0; return 0; } - /* - * Done with the sibling leaf block here, drop the dabuf - * so path_shift can get it. - */ - xfs_da_buf_done(bp); + /* * Make altpath point to the block we want to keep (the lower * numbered block) and path point to the block we want to drop. */ memcpy(&state->altpath, &state->path, sizeof(state->path)); if (blkno < blk->blkno) - error = xfs_da_path_shift(state, &state->altpath, forward, 0, + error = xfs_da3_path_shift(state, &state->altpath, forward, 0, &rval); else - error = xfs_da_path_shift(state, &state->path, forward, 0, + error = xfs_da3_path_shift(state, &state->path, forward, 0, &rval); if (error) { return error; @@ -1243,34 +1551,53 @@ xfs_dir2_leafn_unbalance( xfs_da_args_t *args; /* operation arguments */ xfs_dir2_leaf_t *drop_leaf; /* dead leaf structure */ xfs_dir2_leaf_t *save_leaf; /* surviving leaf structure */ + struct xfs_dir3_icleaf_hdr savehdr; + struct xfs_dir3_icleaf_hdr drophdr; + struct xfs_dir2_leaf_entry *sents; + struct xfs_dir2_leaf_entry *dents; args = state->args; ASSERT(drop_blk->magic == XFS_DIR2_LEAFN_MAGIC); ASSERT(save_blk->magic == XFS_DIR2_LEAFN_MAGIC); - drop_leaf = drop_blk->bp->data; - save_leaf = save_blk->bp->data; - ASSERT(drop_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); - ASSERT(save_leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC)); + drop_leaf = drop_blk->bp->b_addr; + save_leaf = save_blk->bp->b_addr; + + xfs_dir3_leaf_hdr_from_disk(&savehdr, save_leaf); + xfs_dir3_leaf_hdr_from_disk(&drophdr, drop_leaf); + sents = xfs_dir3_leaf_ents_p(save_leaf); + dents = xfs_dir3_leaf_ents_p(drop_leaf); + /* * If there are any stale leaf entries, take this opportunity * to purge them. */ - if (drop_leaf->hdr.stale) - xfs_dir2_leaf_compact(args, drop_blk->bp); - if (save_leaf->hdr.stale) - xfs_dir2_leaf_compact(args, save_blk->bp); + if (drophdr.stale) + xfs_dir3_leaf_compact(args, &drophdr, drop_blk->bp); + if (savehdr.stale) + xfs_dir3_leaf_compact(args, &savehdr, save_blk->bp); + /* * Move the entries from drop to the appropriate end of save. */ - drop_blk->hashval = be32_to_cpu(drop_leaf->ents[be16_to_cpu(drop_leaf->hdr.count) - 1].hashval); + drop_blk->hashval = be32_to_cpu(dents[drophdr.count - 1].hashval); if (xfs_dir2_leafn_order(save_blk->bp, drop_blk->bp)) - xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp, 0, - be16_to_cpu(drop_leaf->hdr.count)); + xfs_dir3_leafn_moveents(args, drop_blk->bp, &drophdr, dents, 0, + save_blk->bp, &savehdr, sents, 0, + drophdr.count); else - xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp, - be16_to_cpu(save_leaf->hdr.count), be16_to_cpu(drop_leaf->hdr.count)); - save_blk->hashval = be32_to_cpu(save_leaf->ents[be16_to_cpu(save_leaf->hdr.count) - 1].hashval); - xfs_dir2_leafn_check(args->dp, save_blk->bp); + xfs_dir3_leafn_moveents(args, drop_blk->bp, &drophdr, dents, 0, + save_blk->bp, &savehdr, sents, + savehdr.count, drophdr.count); + save_blk->hashval = be32_to_cpu(sents[savehdr.count - 1].hashval); + + /* log the changes made when moving the entries */ + xfs_dir3_leaf_hdr_to_disk(save_leaf, &savehdr); + xfs_dir3_leaf_hdr_to_disk(drop_leaf, &drophdr); + xfs_dir3_leaf_log_header(args->trans, save_blk->bp); + xfs_dir3_leaf_log_header(args->trans, drop_blk->bp); + + xfs_dir3_leaf_check(args->dp->i_mount, save_blk->bp); + xfs_dir3_leaf_check(args->dp->i_mount, drop_blk->bp); } /* @@ -1299,7 +1626,7 @@ xfs_dir2_node_addname( * Look up the name. We're not supposed to find it, but * this gives us the insertion point. */ - error = xfs_da_node_lookup_int(state, &rval); + error = xfs_da3_node_lookup_int(state, &rval); if (error) rval = error; if (rval != ENOENT) { @@ -1325,7 +1652,7 @@ xfs_dir2_node_addname( * It worked, fix the hash values up the btree. */ if (!(args->op_flags & XFS_DA_OP_JUSTCHECK)) - xfs_da_fixhashpath(state, &state->path); + xfs_da3_fixhashpath(state, &state->path); } else { /* * It didn't work, we need to split the leaf block. @@ -1337,7 +1664,7 @@ xfs_dir2_node_addname( /* * Split the leaf block and insert the new entry. */ - rval = xfs_da_split(state); + rval = xfs_da3_split(state); } done: xfs_da_state_free(state); @@ -1356,13 +1683,13 @@ xfs_dir2_node_addname_int( { xfs_dir2_data_hdr_t *hdr; /* data block header */ xfs_dir2_db_t dbno; /* data block number */ - xfs_dabuf_t *dbp; /* data block buffer */ + struct xfs_buf *dbp; /* data block buffer */ xfs_dir2_data_entry_t *dep; /* data entry pointer */ xfs_inode_t *dp; /* incore directory inode */ xfs_dir2_data_unused_t *dup; /* data unused entry pointer */ int error; /* error return value */ xfs_dir2_db_t fbno; /* freespace block number */ - xfs_dabuf_t *fbp; /* freespace buffer */ + struct xfs_buf *fbp; /* freespace buffer */ int findex; /* freespace entry index */ xfs_dir2_free_t *free=NULL; /* freespace block structure */ xfs_dir2_db_t ifbno; /* initial freespace block no */ @@ -1374,6 +1701,9 @@ xfs_dir2_node_addname_int( int needscan; /* need to rescan data frees */ __be16 *tagp; /* data entry tag pointer */ xfs_trans_t *tp; /* transaction pointer */ + __be16 *bests; + struct xfs_dir3_icfree_hdr freehdr; + struct xfs_dir2_data_free *bf; dp = args->dp; mp = dp->i_mount; @@ -1390,37 +1720,38 @@ xfs_dir2_node_addname_int( * Remember initial freespace block number. */ ifbno = fblk->blkno; - free = fbp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); + free = fbp->b_addr; findex = fblk->index; + bests = xfs_dir3_free_bests_p(mp, free); + xfs_dir3_free_hdr_from_disk(&freehdr, free); + /* * This means the free entry showed that the data block had * space for our entry, so we remembered it. * Use that data block. */ if (findex >= 0) { - ASSERT(findex < be32_to_cpu(free->hdr.nvalid)); - ASSERT(be16_to_cpu(free->bests[findex]) != NULLDATAOFF); - ASSERT(be16_to_cpu(free->bests[findex]) >= length); - dbno = be32_to_cpu(free->hdr.firstdb) + findex; - } - /* - * The data block looked at didn't have enough room. - * We'll start at the beginning of the freespace entries. - */ - else { + ASSERT(findex < freehdr.nvalid); + ASSERT(be16_to_cpu(bests[findex]) != NULLDATAOFF); + ASSERT(be16_to_cpu(bests[findex]) >= length); + dbno = freehdr.firstdb + findex; + } else { + /* + * The data block looked at didn't have enough room. + * We'll start at the beginning of the freespace entries. + */ dbno = -1; findex = 0; } - } - /* - * Didn't come in with a freespace block, so don't have a data block. - */ - else { + } else { + /* + * Didn't come in with a freespace block, so no data block. + */ ifbno = dbno = -1; fbp = NULL; findex = 0; } + /* * If we don't have a data block yet, we're going to scan the * freespace blocks looking for one. Figure out what the @@ -1466,33 +1797,38 @@ xfs_dir2_node_addname_int( * This should be really rare, so there's no reason * to avoid it. */ - if ((error = xfs_da_read_buf(tp, dp, - xfs_dir2_db_to_da(mp, fbno), -2, &fbp, - XFS_DATA_FORK))) { + error = xfs_dir2_free_try_read(tp, dp, + xfs_dir2_db_to_da(mp, fbno), + &fbp); + if (error) return error; - } - if (unlikely(fbp == NULL)) { + if (!fbp) continue; - } - free = fbp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); + free = fbp->b_addr; findex = 0; } /* * Look at the current free entry. Is it good enough? + * + * The bests initialisation should be where the bufer is read in + * the above branch. But gcc is too stupid to realise that bests + * and the freehdr are actually initialised if they are placed + * there, so we have to do it here to avoid warnings. Blech. */ - if (be16_to_cpu(free->bests[findex]) != NULLDATAOFF && - be16_to_cpu(free->bests[findex]) >= length) - dbno = be32_to_cpu(free->hdr.firstdb) + findex; + bests = xfs_dir3_free_bests_p(mp, free); + xfs_dir3_free_hdr_from_disk(&freehdr, free); + if (be16_to_cpu(bests[findex]) != NULLDATAOFF && + be16_to_cpu(bests[findex]) >= length) + dbno = freehdr.firstdb + findex; else { /* * Are we done with the freeblock? */ - if (++findex == be32_to_cpu(free->hdr.nvalid)) { + if (++findex == freehdr.nvalid) { /* * Drop the block. */ - xfs_da_brelse(tp, fbp); + xfs_trans_brelse(tp, fbp); fbp = NULL; if (fblk && fblk->bp) fblk->bp = NULL; @@ -1507,36 +1843,23 @@ xfs_dir2_node_addname_int( /* * Not allowed to allocate, return failure. */ - if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || - args->total == 0) { - /* - * Drop the freespace buffer unless it came from our - * caller. - */ - if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) - xfs_da_buf_done(fbp); + if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0) return XFS_ERROR(ENOSPC); - } + /* * Allocate and initialize the new data block. */ if (unlikely((error = xfs_dir2_grow_inode(args, XFS_DIR2_DATA_SPACE, &dbno)) || - (error = xfs_dir2_data_init(args, dbno, &dbp)))) { - /* - * Drop the freespace buffer unless it came from our - * caller. - */ - if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) - xfs_da_buf_done(fbp); + (error = xfs_dir3_data_init(args, dbno, &dbp)))) return error; - } + /* * If (somehow) we have a freespace block, get rid of it. */ if (fbp) - xfs_da_brelse(tp, fbp); + xfs_trans_brelse(tp, fbp); if (fblk && fblk->bp) fblk->bp = NULL; @@ -1545,21 +1868,21 @@ xfs_dir2_node_addname_int( * that was just allocated. */ fbno = xfs_dir2_db_to_fdb(mp, dbno); - if (unlikely(error = xfs_da_read_buf(tp, dp, - xfs_dir2_db_to_da(mp, fbno), -2, &fbp, - XFS_DATA_FORK))) { - xfs_da_buf_done(dbp); + error = xfs_dir2_free_try_read(tp, dp, + xfs_dir2_db_to_da(mp, fbno), + &fbp); + if (error) return error; - } + /* * If there wasn't a freespace block, the read will * return a NULL fbp. Allocate and initialize a new one. */ - if( fbp == NULL ) { - if ((error = xfs_dir2_grow_inode(args, XFS_DIR2_FREE_SPACE, - &fbno))) { + if (!fbp) { + error = xfs_dir2_grow_inode(args, XFS_DIR2_FREE_SPACE, + &fbno); + if (error) return error; - } if (unlikely(xfs_dir2_db_to_fdb(mp, dbno) != fbno)) { xfs_alert(mp, @@ -1587,27 +1910,22 @@ xfs_dir2_node_addname_int( /* * Get a buffer for the new block. */ - if ((error = xfs_da_get_buf(tp, dp, - xfs_dir2_db_to_da(mp, fbno), - -1, &fbp, XFS_DATA_FORK))) { + error = xfs_dir3_free_get_buf(tp, dp, fbno, &fbp); + if (error) return error; - } - ASSERT(fbp != NULL); + free = fbp->b_addr; + bests = xfs_dir3_free_bests_p(mp, free); + xfs_dir3_free_hdr_from_disk(&freehdr, free); /* - * Initialize the new block to be empty, and remember - * its first slot as our empty slot. + * Remember the first slot as our empty slot. */ - free = fbp->data; - free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC); - free->hdr.firstdb = cpu_to_be32( - (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) * - xfs_dir2_free_max_bests(mp)); - free->hdr.nvalid = 0; - free->hdr.nused = 0; + freehdr.firstdb = (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) * + xfs_dir3_free_max_bests(mp); } else { - free = fbp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); + free = fbp->b_addr; + bests = xfs_dir3_free_bests_p(mp, free); + xfs_dir3_free_hdr_from_disk(&freehdr, free); } /* @@ -1618,20 +1936,21 @@ xfs_dir2_node_addname_int( * If it's after the end of the current entries in the * freespace block, extend that table. */ - if (findex >= be32_to_cpu(free->hdr.nvalid)) { - ASSERT(findex < xfs_dir2_free_max_bests(mp)); - free->hdr.nvalid = cpu_to_be32(findex + 1); + if (findex >= freehdr.nvalid) { + ASSERT(findex < xfs_dir3_free_max_bests(mp)); + freehdr.nvalid = findex + 1; /* * Tag new entry so nused will go up. */ - free->bests[findex] = cpu_to_be16(NULLDATAOFF); + bests[findex] = cpu_to_be16(NULLDATAOFF); } /* * If this entry was for an empty data block * (this should always be true) then update the header. */ - if (free->bests[findex] == cpu_to_be16(NULLDATAOFF)) { - be32_add_cpu(&free->hdr.nused, 1); + if (bests[findex] == cpu_to_be16(NULLDATAOFF)) { + freehdr.nused++; + xfs_dir3_free_hdr_to_disk(fbp->b_addr, &freehdr); xfs_dir2_free_log_header(tp, fbp); } /* @@ -1639,8 +1958,9 @@ xfs_dir2_node_addname_int( * We haven't allocated the data entry yet so this will * change again. */ - hdr = dbp->data; - free->bests[findex] = hdr->bestfree[0].length; + hdr = dbp->b_addr; + bf = xfs_dir3_data_bestfree_p(hdr); + bests[findex] = bf[0].length; logfree = 1; } /* @@ -1650,30 +1970,26 @@ xfs_dir2_node_addname_int( /* * If just checking, we succeeded. */ - if (args->op_flags & XFS_DA_OP_JUSTCHECK) { - if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) - xfs_da_buf_done(fbp); + if (args->op_flags & XFS_DA_OP_JUSTCHECK) return 0; - } + /* * Read the data block in. */ - if (unlikely( - error = xfs_da_read_buf(tp, dp, xfs_dir2_db_to_da(mp, dbno), - -1, &dbp, XFS_DATA_FORK))) { - if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) - xfs_da_buf_done(fbp); + error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, dbno), + -1, &dbp); + if (error) return error; - } - hdr = dbp->data; + hdr = dbp->b_addr; + bf = xfs_dir3_data_bestfree_p(hdr); logfree = 0; } - ASSERT(be16_to_cpu(hdr->bestfree[0].length) >= length); + ASSERT(be16_to_cpu(bf[0].length) >= length); /* * Point to the existing unused space. */ dup = (xfs_dir2_data_unused_t *) - ((char *)hdr + be16_to_cpu(hdr->bestfree[0].offset)); + ((char *)hdr + be16_to_cpu(bf[0].offset)); needscan = needlog = 0; /* * Mark the first part of the unused space, inuse for us. @@ -1704,8 +2020,9 @@ xfs_dir2_node_addname_int( /* * If the freespace entry is now wrong, update it. */ - if (be16_to_cpu(free->bests[findex]) != be16_to_cpu(hdr->bestfree[0].length)) { - free->bests[findex] = hdr->bestfree[0].length; + bests = xfs_dir3_free_bests_p(mp, free); /* gcc is so stupid */ + if (be16_to_cpu(bests[findex]) != be16_to_cpu(bf[0].length)) { + bests[findex] = bf[0].length; logfree = 1; } /* @@ -1714,22 +2031,16 @@ xfs_dir2_node_addname_int( if (logfree) xfs_dir2_free_log_bests(tp, fbp, findex, findex); /* - * If the caller didn't hand us the freespace block, drop it. - */ - if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL) - xfs_da_buf_done(fbp); - /* * Return the data block and offset in args, then drop the data block. */ args->blkno = (xfs_dablk_t)dbno; args->index = be16_to_cpu(*tagp); - xfs_da_buf_done(dbp); return 0; } /* * Lookup an entry in a node-format directory. - * All the real work happens in xfs_da_node_lookup_int. + * All the real work happens in xfs_da3_node_lookup_int. * The only real output is the inode number of the entry. */ int /* error */ @@ -1754,29 +2065,30 @@ xfs_dir2_node_lookup( /* * Fill in the path to the entry in the cursor. */ - error = xfs_da_node_lookup_int(state, &rval); + error = xfs_da3_node_lookup_int(state, &rval); if (error) rval = error; else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) { /* If a CI match, dup the actual name and return EEXIST */ xfs_dir2_data_entry_t *dep; - dep = (xfs_dir2_data_entry_t *)((char *)state->extrablk.bp-> - data + state->extrablk.index); + dep = (xfs_dir2_data_entry_t *) + ((char *)state->extrablk.bp->b_addr + + state->extrablk.index); rval = xfs_dir_cilookup_result(args, dep->name, dep->namelen); } /* * Release the btree blocks and leaf block. */ for (i = 0; i < state->path.active; i++) { - xfs_da_brelse(args->trans, state->path.blk[i].bp); + xfs_trans_brelse(args->trans, state->path.blk[i].bp); state->path.blk[i].bp = NULL; } /* * Release the data block if we have it. */ if (state->extravalid && state->extrablk.bp) { - xfs_da_brelse(args->trans, state->extrablk.bp); + xfs_trans_brelse(args->trans, state->extrablk.bp); state->extrablk.bp = NULL; } xfs_da_state_free(state); @@ -1808,7 +2120,7 @@ xfs_dir2_node_removename( /* * Look up the entry we're deleting, set up the cursor. */ - error = xfs_da_node_lookup_int(state, &rval); + error = xfs_da3_node_lookup_int(state, &rval); if (error) rval = error; /* @@ -1832,12 +2144,12 @@ xfs_dir2_node_removename( /* * Fix the hash values up the btree. */ - xfs_da_fixhashpath(state, &state->path); + xfs_da3_fixhashpath(state, &state->path); /* * If we need to join leaf blocks, do it. */ if (rval && state->path.active > 1) - error = xfs_da_join(state); + error = xfs_da3_join(state); /* * If no errors so far, try conversion to leaf format. */ @@ -1879,7 +2191,7 @@ xfs_dir2_node_replace( /* * Lookup the entry to change in the btree. */ - error = xfs_da_node_lookup_int(state, &rval); + error = xfs_da3_node_lookup_int(state, &rval); if (error) { rval = error; } @@ -1888,19 +2200,22 @@ xfs_dir2_node_replace( * and locked it. But paranoia is good. */ if (rval == EEXIST) { + struct xfs_dir2_leaf_entry *ents; /* * Find the leaf entry. */ blk = &state->path.blk[state->path.active - 1]; ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC); - leaf = blk->bp->data; - lep = &leaf->ents[blk->index]; + leaf = blk->bp->b_addr; + ents = xfs_dir3_leaf_ents_p(leaf); + lep = &ents[blk->index]; ASSERT(state->extravalid); /* * Point to the data entry. */ - hdr = state->extrablk.bp->data; - ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC)); + hdr = state->extrablk.bp->b_addr; + ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) || + hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC)); dep = (xfs_dir2_data_entry_t *) ((char *)hdr + xfs_dir2_dataptr_to_off(state->mp, be32_to_cpu(lep->address))); @@ -1916,14 +2231,14 @@ xfs_dir2_node_replace( * Didn't find it, and we're holding a data block. Drop it. */ else if (state->extravalid) { - xfs_da_brelse(args->trans, state->extrablk.bp); + xfs_trans_brelse(args->trans, state->extrablk.bp); state->extrablk.bp = NULL; } /* * Release all the buffers in the cursor. */ for (i = 0; i < state->path.active; i++) { - xfs_da_brelse(args->trans, state->path.blk[i].bp); + xfs_trans_brelse(args->trans, state->path.blk[i].bp); state->path.blk[i].bp = NULL; } xfs_da_state_free(state); @@ -1940,12 +2255,13 @@ xfs_dir2_node_trim_free( xfs_fileoff_t fo, /* free block number */ int *rvalp) /* out: did something */ { - xfs_dabuf_t *bp; /* freespace buffer */ + struct xfs_buf *bp; /* freespace buffer */ xfs_inode_t *dp; /* incore directory inode */ int error; /* error return code */ xfs_dir2_free_t *free; /* freespace structure */ xfs_mount_t *mp; /* filesystem mount point */ xfs_trans_t *tp; /* transaction pointer */ + struct xfs_dir3_icfree_hdr freehdr; dp = args->dp; mp = dp->i_mount; @@ -1953,25 +2269,23 @@ xfs_dir2_node_trim_free( /* * Read the freespace block. */ - if (unlikely(error = xfs_da_read_buf(tp, dp, (xfs_dablk_t)fo, -2, &bp, - XFS_DATA_FORK))) { + error = xfs_dir2_free_try_read(tp, dp, fo, &bp); + if (error) return error; - } - /* * There can be holes in freespace. If fo is a hole, there's * nothing to do. */ - if (bp == NULL) { + if (!bp) return 0; - } - free = bp->data; - ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC)); + free = bp->b_addr; + xfs_dir3_free_hdr_from_disk(&freehdr, free); + /* * If there are used entries, there's nothing to do. */ - if (be32_to_cpu(free->hdr.nused) > 0) { - xfs_da_brelse(tp, bp); + if (freehdr.nused > 0) { + xfs_trans_brelse(tp, bp); *rvalp = 0; return 0; } @@ -1987,7 +2301,7 @@ xfs_dir2_node_trim_free( * pieces. This is the last block of an extent. */ ASSERT(error != ENOSPC); - xfs_da_brelse(tp, bp); + xfs_trans_brelse(tp, bp); return error; } /* |