diff options
Diffstat (limited to 'fs/xfs/libxfs')
35 files changed, 1207 insertions, 143 deletions
diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c new file mode 100644 index 000000000000..9345802c99f7 --- /dev/null +++ b/fs/xfs/libxfs/xfs_ag.c @@ -0,0 +1,464 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * Copyright (c) 2018 Red Hat, Inc. + * All rights reserved. + */ + +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_shared.h" +#include "xfs_format.h" +#include "xfs_trans_resv.h" +#include "xfs_sb.h" +#include "xfs_mount.h" +#include "xfs_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_rmap_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_rmap.h" +#include "xfs_ag.h" + +static struct xfs_buf * +xfs_get_aghdr_buf( + struct xfs_mount *mp, + xfs_daddr_t blkno, + size_t numblks, + int flags, + const struct xfs_buf_ops *ops) +{ + struct xfs_buf *bp; + + bp = xfs_buf_get_uncached(mp->m_ddev_targp, numblks, flags); + if (!bp) + return NULL; + + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + bp->b_bn = blkno; + bp->b_maps[0].bm_bn = blkno; + bp->b_ops = ops; + + return bp; +} + +/* + * Generic btree root block init function + */ +static void +xfs_btroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + xfs_btree_init_block(mp, bp, id->type, 0, 0, id->agno, 0); +} + +/* + * Alloc btree root block init functions + */ +static void +xfs_bnoroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_alloc_rec *arec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_BNO, 0, 1, id->agno, 0); + arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); + arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); + arec->ar_blockcount = cpu_to_be32(id->agsize - + be32_to_cpu(arec->ar_startblock)); +} + +static void +xfs_cntroot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_alloc_rec *arec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_CNT, 0, 1, id->agno, 0); + arec = XFS_ALLOC_REC_ADDR(mp, XFS_BUF_TO_BLOCK(bp), 1); + arec->ar_startblock = cpu_to_be32(mp->m_ag_prealloc_blocks); + arec->ar_blockcount = cpu_to_be32(id->agsize - + be32_to_cpu(arec->ar_startblock)); +} + +/* + * Reverse map root block init + */ +static void +xfs_rmaproot_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); + struct xfs_rmap_rec *rrec; + + xfs_btree_init_block(mp, bp, XFS_BTNUM_RMAP, 0, 4, id->agno, 0); + + /* + * mark the AG header regions as static metadata The BNO + * btree block is the first block after the headers, so + * it's location defines the size of region the static + * metadata consumes. + * + * Note: unlike mkfs, we never have to account for log + * space when growing the data regions + */ + rrec = XFS_RMAP_REC_ADDR(block, 1); + rrec->rm_startblock = 0; + rrec->rm_blockcount = cpu_to_be32(XFS_BNO_BLOCK(mp)); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_FS); + rrec->rm_offset = 0; + + /* account freespace btree root blocks */ + rrec = XFS_RMAP_REC_ADDR(block, 2); + rrec->rm_startblock = cpu_to_be32(XFS_BNO_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(2); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); + rrec->rm_offset = 0; + + /* account inode btree root blocks */ + rrec = XFS_RMAP_REC_ADDR(block, 3); + rrec->rm_startblock = cpu_to_be32(XFS_IBT_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(XFS_RMAP_BLOCK(mp) - + XFS_IBT_BLOCK(mp)); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_INOBT); + rrec->rm_offset = 0; + + /* account for rmap btree root */ + rrec = XFS_RMAP_REC_ADDR(block, 4); + rrec->rm_startblock = cpu_to_be32(XFS_RMAP_BLOCK(mp)); + rrec->rm_blockcount = cpu_to_be32(1); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG); + rrec->rm_offset = 0; + + /* account for refc btree root */ + if (xfs_sb_version_hasreflink(&mp->m_sb)) { + rrec = XFS_RMAP_REC_ADDR(block, 5); + rrec->rm_startblock = cpu_to_be32(xfs_refc_block(mp)); + rrec->rm_blockcount = cpu_to_be32(1); + rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_REFC); + rrec->rm_offset = 0; + be16_add_cpu(&block->bb_numrecs, 1); + } +} + +/* + * Initialise new secondary superblocks with the pre-grow geometry, but mark + * them as "in progress" so we know they haven't yet been activated. This will + * get cleared when the update with the new geometry information is done after + * changes to the primary are committed. This isn't strictly necessary, but we + * get it for free with the delayed buffer write lists and it means we can tell + * if a grow operation didn't complete properly after the fact. + */ +static void +xfs_sbblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_dsb *dsb = XFS_BUF_TO_SBP(bp); + + xfs_sb_to_disk(dsb, &mp->m_sb); + dsb->sb_inprogress = 1; +} + +static void +xfs_agfblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agf *agf = XFS_BUF_TO_AGF(bp); + xfs_extlen_t tmpsize; + + agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC); + agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION); + agf->agf_seqno = cpu_to_be32(id->agno); + agf->agf_length = cpu_to_be32(id->agsize); + agf->agf_roots[XFS_BTNUM_BNOi] = cpu_to_be32(XFS_BNO_BLOCK(mp)); + agf->agf_roots[XFS_BTNUM_CNTi] = cpu_to_be32(XFS_CNT_BLOCK(mp)); + agf->agf_levels[XFS_BTNUM_BNOi] = cpu_to_be32(1); + agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1); + if (xfs_sb_version_hasrmapbt(&mp->m_sb)) { + agf->agf_roots[XFS_BTNUM_RMAPi] = + cpu_to_be32(XFS_RMAP_BLOCK(mp)); + agf->agf_levels[XFS_BTNUM_RMAPi] = cpu_to_be32(1); + agf->agf_rmap_blocks = cpu_to_be32(1); + } + + agf->agf_flfirst = cpu_to_be32(1); + agf->agf_fllast = 0; + agf->agf_flcount = 0; + tmpsize = id->agsize - mp->m_ag_prealloc_blocks; + agf->agf_freeblks = cpu_to_be32(tmpsize); + agf->agf_longest = cpu_to_be32(tmpsize); + if (xfs_sb_version_hascrc(&mp->m_sb)) + uuid_copy(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid); + if (xfs_sb_version_hasreflink(&mp->m_sb)) { + agf->agf_refcount_root = cpu_to_be32( + xfs_refc_block(mp)); + agf->agf_refcount_level = cpu_to_be32(1); + agf->agf_refcount_blocks = cpu_to_be32(1); + } +} + +static void +xfs_agflblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp); + __be32 *agfl_bno; + int bucket; + + if (xfs_sb_version_hascrc(&mp->m_sb)) { + agfl->agfl_magicnum = cpu_to_be32(XFS_AGFL_MAGIC); + agfl->agfl_seqno = cpu_to_be32(id->agno); + uuid_copy(&agfl->agfl_uuid, &mp->m_sb.sb_meta_uuid); + } + + agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, bp); + for (bucket = 0; bucket < xfs_agfl_size(mp); bucket++) + agfl_bno[bucket] = cpu_to_be32(NULLAGBLOCK); +} + +static void +xfs_agiblock_init( + struct xfs_mount *mp, + struct xfs_buf *bp, + struct aghdr_init_data *id) +{ + struct xfs_agi *agi = XFS_BUF_TO_AGI(bp); + int bucket; + + agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC); + agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION); + agi->agi_seqno = cpu_to_be32(id->agno); + agi->agi_length = cpu_to_be32(id->agsize); + agi->agi_count = 0; + agi->agi_root = cpu_to_be32(XFS_IBT_BLOCK(mp)); + agi->agi_level = cpu_to_be32(1); + agi->agi_freecount = 0; + agi->agi_newino = cpu_to_be32(NULLAGINO); + agi->agi_dirino = cpu_to_be32(NULLAGINO); + if (xfs_sb_version_hascrc(&mp->m_sb)) + uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid); + if (xfs_sb_version_hasfinobt(&mp->m_sb)) { + agi->agi_free_root = cpu_to_be32(XFS_FIBT_BLOCK(mp)); + agi->agi_free_level = cpu_to_be32(1); + } + for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) + agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); +} + +typedef void (*aghdr_init_work_f)(struct xfs_mount *mp, struct xfs_buf *bp, + struct aghdr_init_data *id); +static int +xfs_ag_init_hdr( + struct xfs_mount *mp, + struct aghdr_init_data *id, + aghdr_init_work_f work, + const struct xfs_buf_ops *ops) + +{ + struct xfs_buf *bp; + + bp = xfs_get_aghdr_buf(mp, id->daddr, id->numblks, 0, ops); + if (!bp) + return -ENOMEM; + + (*work)(mp, bp, id); + + xfs_buf_delwri_queue(bp, &id->buffer_list); + xfs_buf_relse(bp); + return 0; +} + +struct xfs_aghdr_grow_data { + xfs_daddr_t daddr; + size_t numblks; + const struct xfs_buf_ops *ops; + aghdr_init_work_f work; + xfs_btnum_t type; + bool need_init; +}; + +/* + * Prepare new AG headers to be written to disk. We use uncached buffers here, + * as it is assumed these new AG headers are currently beyond the currently + * valid filesystem address space. Using cached buffers would trip over EOFS + * corruption detection alogrithms in the buffer cache lookup routines. + * + * This is a non-transactional function, but the prepared buffers are added to a + * delayed write buffer list supplied by the caller so they can submit them to + * disk and wait on them as required. + */ +int +xfs_ag_init_headers( + struct xfs_mount *mp, + struct aghdr_init_data *id) + +{ + struct xfs_aghdr_grow_data aghdr_data[] = { + { /* SB */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_SB_DADDR), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_sb_buf_ops, + .work = &xfs_sbblock_init, + .need_init = true + }, + { /* AGF */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGF_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agf_buf_ops, + .work = &xfs_agfblock_init, + .need_init = true + }, + { /* AGFL */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGFL_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agfl_buf_ops, + .work = &xfs_agflblock_init, + .need_init = true + }, + { /* AGI */ + .daddr = XFS_AG_DADDR(mp, id->agno, XFS_AGI_DADDR(mp)), + .numblks = XFS_FSS_TO_BB(mp, 1), + .ops = &xfs_agi_buf_ops, + .work = &xfs_agiblock_init, + .need_init = true + }, + { /* BNO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_BNO_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_allocbt_buf_ops, + .work = &xfs_bnoroot_init, + .need_init = true + }, + { /* CNT root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_CNT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_allocbt_buf_ops, + .work = &xfs_cntroot_init, + .need_init = true + }, + { /* INO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_IBT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_inobt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_INO, + .need_init = true + }, + { /* FINO root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_FIBT_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_inobt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_FINO, + .need_init = xfs_sb_version_hasfinobt(&mp->m_sb) + }, + { /* RMAP root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, XFS_RMAP_BLOCK(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_rmapbt_buf_ops, + .work = &xfs_rmaproot_init, + .need_init = xfs_sb_version_hasrmapbt(&mp->m_sb) + }, + { /* REFC root block */ + .daddr = XFS_AGB_TO_DADDR(mp, id->agno, xfs_refc_block(mp)), + .numblks = BTOBB(mp->m_sb.sb_blocksize), + .ops = &xfs_refcountbt_buf_ops, + .work = &xfs_btroot_init, + .type = XFS_BTNUM_REFC, + .need_init = xfs_sb_version_hasreflink(&mp->m_sb) + }, + { /* NULL terminating block */ + .daddr = XFS_BUF_DADDR_NULL, + } + }; + struct xfs_aghdr_grow_data *dp; + int error = 0; + + /* Account for AG free space in new AG */ + id->nfree += id->agsize - mp->m_ag_prealloc_blocks; + for (dp = &aghdr_data[0]; dp->daddr != XFS_BUF_DADDR_NULL; dp++) { + if (!dp->need_init) + continue; + + id->daddr = dp->daddr; + id->numblks = dp->numblks; + id->type = dp->type; + error = xfs_ag_init_hdr(mp, id, dp->work, dp->ops); + if (error) + break; + } + return error; +} + +/* + * Extent the AG indicated by the @id by the length passed in + */ +int +xfs_ag_extend_space( + struct xfs_mount *mp, + struct xfs_trans *tp, + struct aghdr_init_data *id, + xfs_extlen_t len) +{ + struct xfs_owner_info oinfo; + struct xfs_buf *bp; + struct xfs_agi *agi; + struct xfs_agf *agf; + int error; + + /* + * Change the agi length. + */ + error = xfs_ialloc_read_agi(mp, tp, id->agno, &bp); + if (error) + return error; + + agi = XFS_BUF_TO_AGI(bp); + be32_add_cpu(&agi->agi_length, len); + ASSERT(id->agno == mp->m_sb.sb_agcount - 1 || + be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks); + xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH); + + /* + * Change agf length. + */ + error = xfs_alloc_read_agf(mp, tp, id->agno, 0, &bp); + if (error) + return error; + + agf = XFS_BUF_TO_AGF(bp); + be32_add_cpu(&agf->agf_length, len); + ASSERT(agf->agf_length == agi->agi_length); + xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH); + + /* + * Free the new space. + * + * XFS_RMAP_OWN_NULL is used here to tell the rmap btree that + * this doesn't actually exist in the rmap btree. + */ + xfs_rmap_ag_owner(&oinfo, XFS_RMAP_OWN_NULL); + error = xfs_rmap_free(tp, bp, id->agno, + be32_to_cpu(agf->agf_length) - len, + len, &oinfo); + if (error) + return error; + + return xfs_free_extent(tp, XFS_AGB_TO_FSB(mp, id->agno, + be32_to_cpu(agf->agf_length) - len), + len, &oinfo, XFS_AG_RESV_NONE); +} diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h new file mode 100644 index 000000000000..412702e23f61 --- /dev/null +++ b/fs/xfs/libxfs/xfs_ag.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * Copyright (c) 2018 Red Hat, Inc. + * All rights reserved. + */ + +#ifndef __LIBXFS_AG_H +#define __LIBXFS_AG_H 1 + +struct xfs_mount; +struct xfs_trans; + +struct aghdr_init_data { + /* per ag data */ + xfs_agblock_t agno; /* ag to init */ + xfs_extlen_t agsize; /* new AG size */ + struct list_head buffer_list; /* buffer writeback list */ + xfs_rfsblock_t nfree; /* cumulative new free space */ + + /* per header data */ + xfs_daddr_t daddr; /* header location */ + size_t numblks; /* size of header */ + xfs_btnum_t type; /* type of btree root block */ +}; + +int xfs_ag_init_headers(struct xfs_mount *mp, struct aghdr_init_data *id); +int xfs_ag_extend_space(struct xfs_mount *mp, struct xfs_trans *tp, + struct aghdr_init_data *id, xfs_extlen_t len); + +#endif /* __LIBXFS_AG_H */ diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 4bcc095fe44a..dc9dd3805d97 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -39,6 +39,9 @@ #include "xfs_buf_item.h" #include "xfs_log.h" #include "xfs_ag_resv.h" +#include "xfs_bmap.h" + +extern kmem_zone_t *xfs_bmap_free_item_zone; struct workqueue_struct *xfs_alloc_wq; @@ -2060,6 +2063,30 @@ xfs_alloc_space_available( return true; } +int +xfs_free_agfl_block( + struct xfs_trans *tp, + xfs_agnumber_t agno, + xfs_agblock_t agbno, + struct xfs_buf *agbp, + struct xfs_owner_info *oinfo) +{ + int error; + struct xfs_buf *bp; + + error = xfs_free_ag_extent(tp, agbp, agno, agbno, 1, oinfo, + XFS_AG_RESV_AGFL); + if (error) + return error; + + bp = xfs_btree_get_bufs(tp->t_mountp, tp, agno, agbno, 0); + if (!bp) + return -EFSCORRUPTED; + xfs_trans_binval(tp, bp); + + return 0; +} + /* * Check the agfl fields of the agf for inconsistency or corruption. The purpose * is to detect an agfl header padding mismatch between current and early v5 @@ -2148,6 +2175,40 @@ xfs_agfl_reset( } /* + * Defer an AGFL block free. This is effectively equivalent to + * xfs_bmap_add_free() with some special handling particular to AGFL blocks. + * + * Deferring AGFL frees helps prevent log reservation overruns due to too many + * allocation operations in a transaction. AGFL frees are prone to this problem + * because for one they are always freed one at a time. Further, an immediate + * AGFL block free can cause a btree join and require another block free before + * the real allocation can proceed. Deferring the free disconnects freeing up + * the AGFL slot from freeing the block. + */ +STATIC void +xfs_defer_agfl_block( + struct xfs_mount *mp, + struct xfs_defer_ops *dfops, + xfs_agnumber_t agno, + xfs_fsblock_t agbno, + struct xfs_owner_info *oinfo) +{ + struct xfs_extent_free_item *new; /* new element */ + + ASSERT(xfs_bmap_free_item_zone != NULL); + ASSERT(oinfo != NULL); + + new = kmem_zone_alloc(xfs_bmap_free_item_zone, KM_SLEEP); + new->xefi_startblock = XFS_AGB_TO_FSB(mp, agno, agbno); + new->xefi_blockcount = 1; + new->xefi_oinfo = *oinfo; + + trace_xfs_agfl_free_defer(mp, agno, 0, agbno, 1); + + xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_AGFL_FREE, &new->xefi_list); +} + +/* * Decide whether to use this allocation group for this allocation. * If so, fix up the btree freelist's size. */ @@ -2247,21 +2308,20 @@ xfs_alloc_fix_freelist( else xfs_rmap_ag_owner(&targs.oinfo, XFS_RMAP_OWN_AG); while (!(flags & XFS_ALLOC_FLAG_NOSHRINK) && pag->pagf_flcount > need) { - struct xfs_buf *bp; - error = xfs_alloc_get_freelist(tp, agbp, &bno, 0); if (error) goto out_agbp_relse; - error = xfs_free_ag_extent(tp, agbp, args->agno, bno, 1, - &targs.oinfo, XFS_AG_RESV_AGFL); - if (error) - goto out_agbp_relse; - bp = xfs_btree_get_bufs(mp, tp, args->agno, bno, 0); - if (!bp) { - error = -EFSCORRUPTED; - goto out_agbp_relse; + + /* defer agfl frees if dfops is provided */ + if (tp->t_agfl_dfops) { + xfs_defer_agfl_block(mp, tp->t_agfl_dfops, args->agno, + bno, &targs.oinfo); + } else { + error = xfs_free_agfl_block(tp, args->agno, bno, agbp, + &targs.oinfo); + if (error) + goto out_agbp_relse; } - xfs_trans_binval(tp, bp); } targs.tp = tp; @@ -2949,18 +3009,20 @@ out: * after fixing up the freelist. */ int /* error */ -xfs_free_extent( +__xfs_free_extent( struct xfs_trans *tp, /* transaction pointer */ xfs_fsblock_t bno, /* starting block number of extent */ xfs_extlen_t len, /* length of extent */ struct xfs_owner_info *oinfo, /* extent owner */ - enum xfs_ag_resv_type type) /* block reservation type */ + enum xfs_ag_resv_type type, /* block reservation type */ + bool skip_discard) { struct xfs_mount *mp = tp->t_mountp; struct xfs_buf *agbp; xfs_agnumber_t agno = XFS_FSB_TO_AGNO(mp, bno); xfs_agblock_t agbno = XFS_FSB_TO_AGBNO(mp, bno); int error; + unsigned int busy_flags = 0; ASSERT(len != 0); ASSERT(type != XFS_AG_RESV_AGFL); @@ -2984,7 +3046,9 @@ xfs_free_extent( if (error) goto err; - xfs_extent_busy_insert(tp, agno, agbno, len, 0); + if (skip_discard) + busy_flags |= XFS_EXTENT_BUSY_SKIP_DISCARD; + xfs_extent_busy_insert(tp, agno, agbno, len, busy_flags); return 0; err: @@ -3116,3 +3180,40 @@ xfs_alloc_has_record( return xfs_btree_has_record(cur, &low, &high, exists); } + +/* + * Walk all the blocks in the AGFL. The @walk_fn can return any negative + * error code or XFS_BTREE_QUERY_RANGE_ABORT. + */ +int +xfs_agfl_walk( + struct xfs_mount *mp, + struct xfs_agf *agf, + struct xfs_buf *agflbp, + xfs_agfl_walk_fn walk_fn, + void *priv) +{ + __be32 *agfl_bno; + unsigned int i; + int error; + + agfl_bno = XFS_BUF_TO_AGFL_BNO(mp, agflbp); + i = be32_to_cpu(agf->agf_flfirst); + + /* Nothing to walk in an empty AGFL. */ + if (agf->agf_flcount == cpu_to_be32(0)) + return 0; + + /* Otherwise, walk from first to last, wrapping as needed. */ + for (;;) { + error = walk_fn(mp, be32_to_cpu(agfl_bno[i]), priv); + if (error) + return error; + if (i == be32_to_cpu(agf->agf_fllast)) + break; + if (++i == xfs_agfl_size(mp)) + i = 0; + } + + return 0; +} diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index cbf789ea5a4e..0747adcd57d6 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -191,12 +191,24 @@ xfs_alloc_vextent( * Free an extent. */ int /* error */ -xfs_free_extent( +__xfs_free_extent( struct xfs_trans *tp, /* transaction pointer */ xfs_fsblock_t bno, /* starting block number of extent */ xfs_extlen_t len, /* length of extent */ struct xfs_owner_info *oinfo, /* extent owner */ - enum xfs_ag_resv_type type); /* block reservation type */ + enum xfs_ag_resv_type type, /* block reservation type */ + bool skip_discard); + +static inline int +xfs_free_extent( + struct xfs_trans *tp, + xfs_fsblock_t bno, + xfs_extlen_t len, + struct xfs_owner_info *oinfo, + enum xfs_ag_resv_type type) +{ + return __xfs_free_extent(tp, bno, len, oinfo, type, false); +} int /* error */ xfs_alloc_lookup_le( @@ -223,6 +235,8 @@ int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp, xfs_agnumber_t agno, int flags, struct xfs_buf **bpp); int xfs_alloc_read_agfl(struct xfs_mount *mp, struct xfs_trans *tp, xfs_agnumber_t agno, struct xfs_buf **bpp); +int xfs_free_agfl_block(struct xfs_trans *, xfs_agnumber_t, xfs_agblock_t, + struct xfs_buf *, struct xfs_owner_info *); int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags); int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno, struct xfs_buf **agbp); @@ -248,4 +262,9 @@ bool xfs_verify_fsbno(struct xfs_mount *mp, xfs_fsblock_t fsbno); int xfs_alloc_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, bool *exist); +typedef int (*xfs_agfl_walk_fn)(struct xfs_mount *mp, xfs_agblock_t bno, + void *priv); +int xfs_agfl_walk(struct xfs_mount *mp, struct xfs_agf *agf, + struct xfs_buf *agflbp, xfs_agfl_walk_fn walk_fn, void *priv); + #endif /* __XFS_ALLOC_H__ */ diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c index b451649ba176..18aec7a0e599 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.c +++ b/fs/xfs/libxfs/xfs_alloc_btree.c @@ -547,3 +547,12 @@ xfs_allocbt_maxrecs( return blocklen / sizeof(xfs_alloc_rec_t); return blocklen / (sizeof(xfs_alloc_key_t) + sizeof(xfs_alloc_ptr_t)); } + +/* Calculate the freespace btree size for some records. */ +xfs_extlen_t +xfs_allocbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_alloc_mnr, len); +} diff --git a/fs/xfs/libxfs/xfs_alloc_btree.h b/fs/xfs/libxfs/xfs_alloc_btree.h index 45e189e7e81c..2fd54728871c 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.h +++ b/fs/xfs/libxfs/xfs_alloc_btree.h @@ -61,5 +61,7 @@ extern struct xfs_btree_cur *xfs_allocbt_init_cursor(struct xfs_mount *, struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t, xfs_btnum_t); extern int xfs_allocbt_maxrecs(struct xfs_mount *, int, int); +extern xfs_extlen_t xfs_allocbt_calc_size(struct xfs_mount *mp, + unsigned long long len); #endif /* __XFS_ALLOC_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index ce4a34a2751d..c3d02a66d39d 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -236,7 +236,7 @@ xfs_attr_set( args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT; args.total = xfs_attr_calc_size(&args, &local); - error = xfs_qm_dqattach(dp, 0); + error = xfs_qm_dqattach(dp); if (error) return error; @@ -427,7 +427,7 @@ xfs_attr_remove( */ args.op_flags = XFS_DA_OP_OKNOENT; - error = xfs_qm_dqattach(dp, 0); + error = xfs_qm_dqattach(dp); if (error) return error; @@ -511,7 +511,14 @@ xfs_attr_shortform_addname(xfs_da_args_t *args) if (args->flags & ATTR_CREATE) return retval; retval = xfs_attr_shortform_remove(args); - ASSERT(retval == 0); + if (retval) + return retval; + /* + * Since we have removed the old attr, clear ATTR_REPLACE so + * that the leaf format add routine won't trip over the attr + * not being around. + */ + args->flags &= ~ATTR_REPLACE; } if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX || diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 21be186067a2..83a6d3c7f872 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -620,7 +620,7 @@ xfs_attr_rmtval_remove( /* * If the "remote" value is in the cache, remove it. */ - bp = xfs_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK); + bp = xfs_buf_incore(mp->m_ddev_targp, dblkno, dblkcnt, XBF_TRYLOCK); if (bp) { xfs_buf_stale(bp); xfs_buf_relse(bp); diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 6a7c2f03ea11..7b0e2b551e23 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -246,7 +246,7 @@ xfs_bmap_get_bp( struct xfs_btree_cur *cur, xfs_fsblock_t bno) { - struct xfs_log_item_desc *lidp; + struct xfs_log_item *lip; int i; if (!cur) @@ -260,9 +260,9 @@ xfs_bmap_get_bp( } /* Chase down all the log items to see if the bp is there */ - list_for_each_entry(lidp, &cur->bc_tp->t_items, lid_trans) { - struct xfs_buf_log_item *bip; - bip = (struct xfs_buf_log_item *)lidp->lid_item; + list_for_each_entry(lip, &cur->bc_tp->t_items, li_trans) { + struct xfs_buf_log_item *bip = (struct xfs_buf_log_item *)lip; + if (bip->bli_item.li_type == XFS_LI_BUF && XFS_BUF_ADDR(bip->bli_buf) == bno) return bip->bli_buf; @@ -312,8 +312,9 @@ xfs_check_block( xfs_warn(mp, "%s: thispa(%d) == pp(%d) %Ld", __func__, j, i, (unsigned long long)be64_to_cpu(*thispa)); - panic("%s: ptrs are equal in node\n", + xfs_err(mp, "%s: ptrs are equal in node\n", __func__); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); } } } @@ -483,7 +484,8 @@ error0: error_norelse: xfs_warn(mp, "%s: BAD after btree leaves for %d extents", __func__, i); - panic("%s: CORRUPTED BTREE OR SOMETHING", __func__); + xfs_err(mp, "%s: CORRUPTED BTREE OR SOMETHING", __func__); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return; } @@ -542,12 +544,13 @@ xfs_bmap_validate_ret( * The list is maintained sorted (by block number). */ void -xfs_bmap_add_free( +__xfs_bmap_add_free( struct xfs_mount *mp, struct xfs_defer_ops *dfops, xfs_fsblock_t bno, xfs_filblks_t len, - struct xfs_owner_info *oinfo) + struct xfs_owner_info *oinfo, + bool skip_discard) { struct xfs_extent_free_item *new; /* new element */ #ifdef DEBUG @@ -574,6 +577,7 @@ xfs_bmap_add_free( new->xefi_oinfo = *oinfo; else xfs_rmap_skip_owner_update(&new->xefi_oinfo); + new->xefi_skip_discard = skip_discard; trace_xfs_bmap_free_defer(mp, XFS_FSB_TO_AGNO(mp, bno), 0, XFS_FSB_TO_AGBNO(mp, bno), len); xfs_defer_add(dfops, XFS_DEFER_OPS_TYPE_FREE, &new->xefi_list); @@ -725,12 +729,16 @@ xfs_bmap_extents_to_btree( *logflagsp = 0; if ((error = xfs_alloc_vextent(&args))) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return error; } if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) { xfs_iroot_realloc(ip, -1, whichfork); + ASSERT(ifp->if_broot == NULL); + XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); xfs_btree_del_cursor(cur, XFS_BTREE_ERROR); return -ENOSPC; } @@ -1997,10 +2005,13 @@ xfs_bmap_add_extent_delay_real( ASSERT(0); } - /* add reverse mapping */ - error = xfs_rmap_map_extent(mp, bma->dfops, bma->ip, whichfork, new); - if (error) - goto done; + /* add reverse mapping unless caller opted out */ + if (!(bma->flags & XFS_BMAPI_NORMAP)) { + error = xfs_rmap_map_extent(mp, bma->dfops, bma->ip, + whichfork, new); + if (error) + goto done; + } /* convert to a btree if necessary */ if (xfs_bmap_needs_btree(bma->ip, whichfork)) { @@ -2664,7 +2675,8 @@ xfs_bmap_add_extent_hole_real( struct xfs_bmbt_irec *new, xfs_fsblock_t *first, struct xfs_defer_ops *dfops, - int *logflagsp) + int *logflagsp, + int flags) { struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); struct xfs_mount *mp = ip->i_mount; @@ -2841,10 +2853,12 @@ xfs_bmap_add_extent_hole_real( break; } - /* add reverse mapping */ - error = xfs_rmap_map_extent(mp, dfops, ip, whichfork, new); - if (error) - goto done; + /* add reverse mapping unless caller opted out */ + if (!(flags & XFS_BMAPI_NORMAP)) { + error = xfs_rmap_map_extent(mp, dfops, ip, whichfork, new); + if (error) + goto done; + } /* convert to a btree if necessary */ if (xfs_bmap_needs_btree(ip, whichfork)) { @@ -4119,7 +4133,8 @@ xfs_bmapi_allocate( else error = xfs_bmap_add_extent_hole_real(bma->tp, bma->ip, whichfork, &bma->icur, &bma->cur, &bma->got, - bma->firstblock, bma->dfops, &bma->logflags); + bma->firstblock, bma->dfops, &bma->logflags, + bma->flags); bma->logflags |= tmp_logflags; if (error) @@ -4505,30 +4520,37 @@ error0: return error; } -static int +int xfs_bmapi_remap( struct xfs_trans *tp, struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len, xfs_fsblock_t startblock, - struct xfs_defer_ops *dfops) + struct xfs_defer_ops *dfops, + int flags) { struct xfs_mount *mp = ip->i_mount; - struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); + struct xfs_ifork *ifp; struct xfs_btree_cur *cur = NULL; xfs_fsblock_t firstblock = NULLFSBLOCK; struct xfs_bmbt_irec got; struct xfs_iext_cursor icur; + int whichfork = xfs_bmapi_whichfork(flags); int logflags = 0, error; + ifp = XFS_IFORK_PTR(ip, whichfork); ASSERT(len > 0); ASSERT(len <= (xfs_filblks_t)MAXEXTLEN); ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(!(flags & ~(XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC | + XFS_BMAPI_NORMAP))); + ASSERT((flags & (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)) != + (XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC)); if (unlikely(XFS_TEST_ERROR( - (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS && - XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE), + (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS && + XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE), mp, XFS_ERRTAG_BMAPIFORMAT))) { XFS_ERROR_REPORT("xfs_bmapi_remap", XFS_ERRLEVEL_LOW, mp); return -EFSCORRUPTED; @@ -4538,7 +4560,7 @@ xfs_bmapi_remap( return -EIO; if (!(ifp->if_flags & XFS_IFEXTENTS)) { - error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK); + error = xfs_iread_extents(tp, ip, whichfork); if (error) return error; } @@ -4553,7 +4575,7 @@ xfs_bmapi_remap( xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); if (ifp->if_flags & XFS_IFBROOT) { - cur = xfs_bmbt_init_cursor(mp, tp, ip, XFS_DATA_FORK); + cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork); cur->bc_private.b.firstblock = firstblock; cur->bc_private.b.dfops = dfops; cur->bc_private.b.flags = 0; @@ -4562,18 +4584,21 @@ xfs_bmapi_remap( got.br_startoff = bno; got.br_startblock = startblock; got.br_blockcount = len; - got.br_state = XFS_EXT_NORM; + if (flags & XFS_BMAPI_PREALLOC) + got.br_state = XFS_EXT_UNWRITTEN; + else + got.br_state = XFS_EXT_NORM; - error = xfs_bmap_add_extent_hole_real(tp, ip, XFS_DATA_FORK, &icur, - &cur, &got, &firstblock, dfops, &logflags); + error = xfs_bmap_add_extent_hole_real(tp, ip, whichfork, &icur, + &cur, &got, &firstblock, dfops, &logflags, flags); if (error) goto error0; - if (xfs_bmap_wants_extents(ip, XFS_DATA_FORK)) { + if (xfs_bmap_wants_extents(ip, whichfork)) { int tmp_logflags = 0; error = xfs_bmap_btree_to_extents(tp, ip, cur, - &tmp_logflags, XFS_DATA_FORK); + &tmp_logflags, whichfork); logflags |= tmp_logflags; } @@ -5100,9 +5125,12 @@ xfs_bmap_del_extent_real( error = xfs_refcount_decrease_extent(mp, dfops, del); if (error) goto done; - } else - xfs_bmap_add_free(mp, dfops, del->br_startblock, - del->br_blockcount, NULL); + } else { + __xfs_bmap_add_free(mp, dfops, del->br_startblock, + del->br_blockcount, NULL, + (bflags & XFS_BMAPI_NODISCARD) || + del->br_state == XFS_EXT_UNWRITTEN); + } } /* @@ -6144,7 +6172,7 @@ xfs_bmap_finish_one( switch (type) { case XFS_BMAP_MAP: error = xfs_bmapi_remap(tp, ip, startoff, *blockcount, - startblock, dfops); + startblock, dfops, 0); *blockcount = 0; break; case XFS_BMAP_UNMAP: diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index 2b766b37096d..2c233f9f1a26 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -68,6 +68,7 @@ struct xfs_extent_free_item xfs_extlen_t xefi_blockcount;/* number of blocks in extent */ struct list_head xefi_list; struct xfs_owner_info xefi_oinfo; /* extent owner */ + bool xefi_skip_discard; }; #define XFS_BMAP_MAX_NMAP 4 @@ -116,6 +117,12 @@ struct xfs_extent_free_item /* Only convert unwritten extents, don't allocate new blocks */ #define XFS_BMAPI_CONVERT_ONLY 0x800 +/* Skip online discard of freed extents */ +#define XFS_BMAPI_NODISCARD 0x1000 + +/* Do not update the rmap btree. Used for reconstructing bmbt from rmapbt. */ +#define XFS_BMAPI_NORMAP 0x2000 + #define XFS_BMAPI_FLAGS \ { XFS_BMAPI_ENTIRE, "ENTIRE" }, \ { XFS_BMAPI_METADATA, "METADATA" }, \ @@ -128,7 +135,9 @@ struct xfs_extent_free_item { XFS_BMAPI_REMAP, "REMAP" }, \ { XFS_BMAPI_COWFORK, "COWFORK" }, \ { XFS_BMAPI_DELALLOC, "DELALLOC" }, \ - { XFS_BMAPI_CONVERT_ONLY, "CONVERT_ONLY" } + { XFS_BMAPI_CONVERT_ONLY, "CONVERT_ONLY" }, \ + { XFS_BMAPI_NODISCARD, "NODISCARD" }, \ + { XFS_BMAPI_NORMAP, "NORMAP" } static inline int xfs_bmapi_aflag(int w) @@ -192,9 +201,9 @@ void xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno, void xfs_trim_extent_eof(struct xfs_bmbt_irec *, struct xfs_inode *); int xfs_bmap_add_attrfork(struct xfs_inode *ip, int size, int rsvd); void xfs_bmap_local_to_extents_empty(struct xfs_inode *ip, int whichfork); -void xfs_bmap_add_free(struct xfs_mount *mp, struct xfs_defer_ops *dfops, +void __xfs_bmap_add_free(struct xfs_mount *mp, struct xfs_defer_ops *dfops, xfs_fsblock_t bno, xfs_filblks_t len, - struct xfs_owner_info *oinfo); + struct xfs_owner_info *oinfo, bool skip_discard); void xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork); int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip, xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork); @@ -240,6 +249,17 @@ int xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork, struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur, int eof); +static inline void +xfs_bmap_add_free( + struct xfs_mount *mp, + struct xfs_defer_ops *dfops, + xfs_fsblock_t bno, + xfs_filblks_t len, + struct xfs_owner_info *oinfo) +{ + __xfs_bmap_add_free(mp, dfops, bno, len, oinfo, false); +} + enum xfs_bmap_intent_type { XFS_BMAP_MAP = 1, XFS_BMAP_UNMAP, @@ -277,4 +297,8 @@ static inline int xfs_bmap_fork_to_state(int whichfork) xfs_failaddr_t xfs_bmap_validate_extent(struct xfs_inode *ip, int whichfork, struct xfs_bmbt_irec *irec); +int xfs_bmapi_remap(struct xfs_trans *tp, struct xfs_inode *ip, + xfs_fileoff_t bno, xfs_filblks_t len, xfs_fsblock_t startblock, + struct xfs_defer_ops *dfops, int flags); + #endif /* __XFS_BMAP_H__ */ diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index d89d06bea6e3..ac9d4aeedb09 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -660,3 +660,12 @@ xfs_bmbt_change_owner( xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); return error; } + +/* Calculate the bmap btree size for some records. */ +unsigned long long +xfs_bmbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_bmap_dmnr, len); +} diff --git a/fs/xfs/libxfs/xfs_bmap_btree.h b/fs/xfs/libxfs/xfs_bmap_btree.h index e4505746ccaa..fb3cd2d9e0f8 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.h +++ b/fs/xfs/libxfs/xfs_bmap_btree.h @@ -118,4 +118,7 @@ extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip, extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *, struct xfs_trans *, struct xfs_inode *, int); +extern unsigned long long xfs_bmbt_calc_size(struct xfs_mount *mp, + unsigned long long len); + #endif /* __XFS_BMAP_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index ac7d66427e42..c825c8182b30 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -4836,14 +4836,14 @@ xfs_btree_query_all( * Calculate the number of blocks needed to store a given number of records * in a short-format (per-AG metadata) btree. */ -xfs_extlen_t +unsigned long long xfs_btree_calc_size( uint *limits, unsigned long long len) { int level; int maxrecs; - xfs_extlen_t rval; + unsigned long long rval; maxrecs = limits[0]; for (level = 0, rval = 0; len > 1; level++) { @@ -4919,3 +4919,24 @@ xfs_btree_has_record( *exists = false; return error; } + +/* Are there more records in this btree? */ +bool +xfs_btree_has_more_records( + struct xfs_btree_cur *cur) +{ + struct xfs_btree_block *block; + struct xfs_buf *bp; + + block = xfs_btree_get_block(cur, 0, &bp); + + /* There are still records in this block. */ + if (cur->bc_ptrs[0] < xfs_btree_get_numrecs(block)) + return true; + + /* There are more record blocks. */ + if (cur->bc_flags & XFS_BTREE_LONG_PTRS) + return block->bb_u.l.bb_rightsib != cpu_to_be64(NULLFSBLOCK); + else + return block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK); +} diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h index 9227159a751e..d7911efee6dc 100644 --- a/fs/xfs/libxfs/xfs_btree.h +++ b/fs/xfs/libxfs/xfs_btree.h @@ -482,7 +482,7 @@ xfs_failaddr_t xfs_btree_lblock_verify(struct xfs_buf *bp, unsigned int max_recs); uint xfs_btree_compute_maxlevels(uint *limits, unsigned long len); -xfs_extlen_t xfs_btree_calc_size(uint *limits, unsigned long long len); +unsigned long long xfs_btree_calc_size(uint *limits, unsigned long long len); /* return codes */ #define XFS_BTREE_QUERY_RANGE_CONTINUE 0 /* keep iterating */ @@ -528,5 +528,6 @@ union xfs_btree_key *xfs_btree_high_key_from_key(struct xfs_btree_cur *cur, union xfs_btree_key *key); int xfs_btree_has_record(struct xfs_btree_cur *cur, union xfs_btree_irec *low, union xfs_btree_irec *high, bool *exists); +bool xfs_btree_has_more_records(struct xfs_btree_cur *cur); #endif /* __XFS_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c index 087fea02c389..3daf175e2535 100644 --- a/fs/xfs/libxfs/xfs_defer.c +++ b/fs/xfs/libxfs/xfs_defer.c @@ -220,7 +220,7 @@ xfs_defer_trans_abort( { struct xfs_defer_pending *dfp; - trace_xfs_defer_trans_abort(tp->t_mountp, dop); + trace_xfs_defer_trans_abort(tp->t_mountp, dop, _RET_IP_); /* Abort intent items that don't have a done item. */ list_for_each_entry(dfp, &dop->dop_pending, dfp_list) { @@ -253,7 +253,7 @@ xfs_defer_trans_roll( for (i = 0; i < XFS_DEFER_OPS_NR_BUFS && dop->dop_bufs[i]; i++) xfs_trans_dirty_buf(*tp, dop->dop_bufs[i]); - trace_xfs_defer_trans_roll((*tp)->t_mountp, dop); + trace_xfs_defer_trans_roll((*tp)->t_mountp, dop, _RET_IP_); /* Roll the transaction. */ error = xfs_trans_roll(tp); @@ -352,10 +352,21 @@ xfs_defer_finish( void *state; int error = 0; void (*cleanup_fn)(struct xfs_trans *, void *, int); + struct xfs_defer_ops *orig_dop; ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); - trace_xfs_defer_finish((*tp)->t_mountp, dop); + trace_xfs_defer_finish((*tp)->t_mountp, dop, _RET_IP_); + + /* + * Attach dfops to the transaction during deferred ops processing. This + * explicitly causes calls into the allocator to defer AGFL block frees. + * Note that this code can go away once all dfops users attach to the + * associated tp. + */ + ASSERT(!(*tp)->t_agfl_dfops || ((*tp)->t_agfl_dfops == dop)); + orig_dop = (*tp)->t_agfl_dfops; + (*tp)->t_agfl_dfops = dop; /* Until we run out of pending work to finish... */ while (xfs_defer_has_unfinished_work(dop)) { @@ -428,10 +439,11 @@ xfs_defer_finish( } out: + (*tp)->t_agfl_dfops = orig_dop; if (error) trace_xfs_defer_finish_error((*tp)->t_mountp, dop, error); else - trace_xfs_defer_finish_done((*tp)->t_mountp, dop); + trace_xfs_defer_finish_done((*tp)->t_mountp, dop, _RET_IP_); return error; } @@ -447,7 +459,7 @@ xfs_defer_cancel( struct list_head *pwi; struct list_head *n; - trace_xfs_defer_cancel(NULL, dop); + trace_xfs_defer_cancel(NULL, dop, _RET_IP_); /* * Free the pending items. Caller should already have arranged @@ -532,5 +544,5 @@ xfs_defer_init( *fbp = NULLFSBLOCK; INIT_LIST_HEAD(&dop->dop_intake); INIT_LIST_HEAD(&dop->dop_pending); - trace_xfs_defer_init(NULL, dop); + trace_xfs_defer_init(NULL, dop, _RET_IP_); } diff --git a/fs/xfs/libxfs/xfs_defer.h b/fs/xfs/libxfs/xfs_defer.h index 045beacdd37d..e70725ba1f5f 100644 --- a/fs/xfs/libxfs/xfs_defer.h +++ b/fs/xfs/libxfs/xfs_defer.h @@ -55,6 +55,7 @@ enum xfs_defer_ops_type { XFS_DEFER_OPS_TYPE_REFCOUNT, XFS_DEFER_OPS_TYPE_RMAP, XFS_DEFER_OPS_TYPE_FREE, + XFS_DEFER_OPS_TYPE_AGFL_FREE, XFS_DEFER_OPS_TYPE_MAX, }; diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c index 8b7a6c3cb599..cce520becee4 100644 --- a/fs/xfs/libxfs/xfs_dquot_buf.c +++ b/fs/xfs/libxfs/xfs_dquot_buf.c @@ -41,14 +41,18 @@ xfs_calc_dquots_per_chunk( /* * Do some primitive error checking on ondisk dquot data structures. + * + * The xfs_dqblk structure /contains/ the xfs_disk_dquot structure; + * we verify them separately because at some points we have only the + * smaller xfs_disk_dquot structure available. */ + xfs_failaddr_t xfs_dquot_verify( struct xfs_mount *mp, xfs_disk_dquot_t *ddq, xfs_dqid_t id, - uint type, /* used only when IO_dorepair is true */ - uint flags) + uint type) /* used only during quotacheck */ { /* * We can encounter an uninitialized dquot buffer for 2 reasons: @@ -70,6 +74,8 @@ xfs_dquot_verify( if (ddq->d_version != XFS_DQUOT_VERSION) return __this_address; + if (type && ddq->d_flags != type) + return __this_address; if (ddq->d_flags != XFS_DQ_USER && ddq->d_flags != XFS_DQ_PROJ && ddq->d_flags != XFS_DQ_GROUP) @@ -99,33 +105,44 @@ xfs_dquot_verify( return NULL; } +xfs_failaddr_t +xfs_dqblk_verify( + struct xfs_mount *mp, + struct xfs_dqblk *dqb, + xfs_dqid_t id, + uint type) /* used only during quotacheck */ +{ + if (xfs_sb_version_hascrc(&mp->m_sb) && + !uuid_equal(&dqb->dd_uuid, &mp->m_sb.sb_meta_uuid)) + return __this_address; + + return xfs_dquot_verify(mp, &dqb->dd_diskdq, id, type); +} + /* * Do some primitive error checking on ondisk dquot data structures. */ int -xfs_dquot_repair( +xfs_dqblk_repair( struct xfs_mount *mp, - struct xfs_disk_dquot *ddq, + struct xfs_dqblk *dqb, xfs_dqid_t id, uint type) { - struct xfs_dqblk *d = (struct xfs_dqblk *)ddq; - - /* * Typically, a repair is only requested by quotacheck. */ ASSERT(id != -1); - memset(d, 0, sizeof(xfs_dqblk_t)); + memset(dqb, 0, sizeof(xfs_dqblk_t)); - d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); - d->dd_diskdq.d_version = XFS_DQUOT_VERSION; - d->dd_diskdq.d_flags = type; - d->dd_diskdq.d_id = cpu_to_be32(id); + dqb->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + dqb->dd_diskdq.d_version = XFS_DQUOT_VERSION; + dqb->dd_diskdq.d_flags = type; + dqb->dd_diskdq.d_id = cpu_to_be32(id); if (xfs_sb_version_hascrc(&mp->m_sb)) { - uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid); - xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk), + uuid_copy(&dqb->dd_uuid, &mp->m_sb.sb_meta_uuid); + xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk), XFS_DQUOT_CRC_OFF); } @@ -135,7 +152,8 @@ xfs_dquot_repair( STATIC bool xfs_dquot_buf_verify_crc( struct xfs_mount *mp, - struct xfs_buf *bp) + struct xfs_buf *bp, + bool readahead) { struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; int ndquots; @@ -156,10 +174,12 @@ xfs_dquot_buf_verify_crc( for (i = 0; i < ndquots; i++, d++) { if (!xfs_verify_cksum((char *)d, sizeof(struct xfs_dqblk), - XFS_DQUOT_CRC_OFF)) - return false; - if (!uuid_equal(&d->dd_uuid, &mp->m_sb.sb_meta_uuid)) + XFS_DQUOT_CRC_OFF)) { + if (!readahead) + xfs_buf_verifier_error(bp, -EFSBADCRC, __func__, + d, sizeof(*d), __this_address); return false; + } } return true; } @@ -167,9 +187,10 @@ xfs_dquot_buf_verify_crc( STATIC xfs_failaddr_t xfs_dquot_buf_verify( struct xfs_mount *mp, - struct xfs_buf *bp) + struct xfs_buf *bp, + bool readahead) { - struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; + struct xfs_dqblk *dqb = bp->b_addr; xfs_failaddr_t fa; xfs_dqid_t id = 0; int ndquots; @@ -195,14 +216,19 @@ xfs_dquot_buf_verify( for (i = 0; i < ndquots; i++) { struct xfs_disk_dquot *ddq; - ddq = &d[i].dd_diskdq; + ddq = &dqb[i].dd_diskdq; if (i == 0) id = be32_to_cpu(ddq->d_id); - fa = xfs_dquot_verify(mp, ddq, id + i, 0, 0); - if (fa) + fa = xfs_dqblk_verify(mp, &dqb[i], id + i, 0); + if (fa) { + if (!readahead) + xfs_buf_verifier_error(bp, -EFSCORRUPTED, + __func__, &dqb[i], + sizeof(struct xfs_dqblk), fa); return fa; + } } return NULL; @@ -214,7 +240,7 @@ xfs_dquot_buf_verify_struct( { struct xfs_mount *mp = bp->b_target->bt_mount; - return xfs_dquot_buf_verify(mp, bp); + return xfs_dquot_buf_verify(mp, bp, false); } static void @@ -222,15 +248,10 @@ xfs_dquot_buf_read_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - xfs_failaddr_t fa; - if (!xfs_dquot_buf_verify_crc(mp, bp)) - xfs_verifier_error(bp, -EFSBADCRC, __this_address); - else { - fa = xfs_dquot_buf_verify(mp, bp); - if (fa) - xfs_verifier_error(bp, -EFSCORRUPTED, __this_address); - } + if (!xfs_dquot_buf_verify_crc(mp, bp, false)) + return; + xfs_dquot_buf_verify(mp, bp, false); } /* @@ -245,8 +266,8 @@ xfs_dquot_buf_readahead_verify( { struct xfs_mount *mp = bp->b_target->bt_mount; - if (!xfs_dquot_buf_verify_crc(mp, bp) || - xfs_dquot_buf_verify(mp, bp) != NULL) { + if (!xfs_dquot_buf_verify_crc(mp, bp, true) || + xfs_dquot_buf_verify(mp, bp, true) != NULL) { xfs_buf_ioerror(bp, -EIO); bp->b_flags &= ~XBF_DONE; } @@ -262,11 +283,8 @@ xfs_dquot_buf_write_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - xfs_failaddr_t fa; - fa = xfs_dquot_buf_verify(mp, bp); - if (fa) - xfs_verifier_error(bp, -EFSCORRUPTED, __this_address); + xfs_dquot_buf_verify(mp, bp, false); } const struct xfs_buf_ops xfs_dquot_buf_ops = { diff --git a/fs/xfs/libxfs/xfs_errortag.h b/fs/xfs/libxfs/xfs_errortag.h index bc1789d95152..d47b91625945 100644 --- a/fs/xfs/libxfs/xfs_errortag.h +++ b/fs/xfs/libxfs/xfs_errortag.h @@ -65,7 +65,8 @@ #define XFS_ERRTAG_LOG_BAD_CRC 29 #define XFS_ERRTAG_LOG_ITEM_PIN 30 #define XFS_ERRTAG_BUF_LRU_REF 31 -#define XFS_ERRTAG_MAX 32 +#define XFS_ERRTAG_FORCE_SCRUB_REPAIR 32 +#define XFS_ERRTAG_MAX 33 /* * Random factors for above tags, 1 means always, 2 means 1/2 time, etc. @@ -102,5 +103,6 @@ #define XFS_RANDOM_LOG_BAD_CRC 1 #define XFS_RANDOM_LOG_ITEM_PIN 1 #define XFS_RANDOM_BUF_LRU_REF 2 +#define XFS_RANDOM_FORCE_SCRUB_REPAIR 1 #endif /* __XFS_ERRORTAG_H_ */ diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index 42956d8d95ed..c1cb29a5f4f6 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -98,6 +98,9 @@ struct xfs_ifork; XFS_SB_VERSION2_PROJID32BIT | \ XFS_SB_VERSION2_FTYPE) +/* Maximum size of the xfs filesystem label, no terminating NULL */ +#define XFSLABEL_MAX 12 + /* * Superblock - in core version. Must match the ondisk version below. * Must be padded to 64 bit alignment. @@ -122,7 +125,7 @@ typedef struct xfs_sb { uint16_t sb_sectsize; /* volume sector size, bytes */ uint16_t sb_inodesize; /* inode size, bytes */ uint16_t sb_inopblock; /* inodes per block */ - char sb_fname[12]; /* file system name */ + char sb_fname[XFSLABEL_MAX]; /* file system name */ uint8_t sb_blocklog; /* log2 of sb_blocksize */ uint8_t sb_sectlog; /* log2 of sb_sectsize */ uint8_t sb_inodelog; /* log2 of sb_inodesize */ @@ -213,7 +216,7 @@ typedef struct xfs_dsb { __be16 sb_sectsize; /* volume sector size, bytes */ __be16 sb_inodesize; /* inode size, bytes */ __be16 sb_inopblock; /* inodes per block */ - char sb_fname[12]; /* file system name */ + char sb_fname[XFSLABEL_MAX]; /* file system name */ __u8 sb_blocklog; /* log2 of sb_blocksize */ __u8 sb_sectlog; /* log2 of sb_sectsize */ __u8 sb_inodelog; /* log2 of sb_inodesize */ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index faf1a4edd618..dddc75e4f1f6 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -542,13 +542,20 @@ struct xfs_scrub_metadata { /* o: Metadata object looked funny but isn't corrupt. */ #define XFS_SCRUB_OFLAG_WARNING (1 << 6) +/* + * o: IFLAG_REPAIR was set but metadata object did not need fixing or + * optimization and has therefore not been altered. + */ +#define XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED (1 << 7) + #define XFS_SCRUB_FLAGS_IN (XFS_SCRUB_IFLAG_REPAIR) #define XFS_SCRUB_FLAGS_OUT (XFS_SCRUB_OFLAG_CORRUPT | \ XFS_SCRUB_OFLAG_PREEN | \ XFS_SCRUB_OFLAG_XFAIL | \ XFS_SCRUB_OFLAG_XCORRUPT | \ XFS_SCRUB_OFLAG_INCOMPLETE | \ - XFS_SCRUB_OFLAG_WARNING) + XFS_SCRUB_OFLAG_WARNING | \ + XFS_SCRUB_OFLAG_NO_REPAIR_NEEDED) #define XFS_SCRUB_FLAGS_ALL (XFS_SCRUB_FLAGS_IN | XFS_SCRUB_FLAGS_OUT) /* diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index de627fa19168..4ca4ff7a757d 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -148,7 +148,7 @@ xfs_inobt_get_rec( /* * Insert a single inobt record. Cursor must already point to desired location. */ -STATIC int +int xfs_inobt_insert_rec( struct xfs_btree_cur *cur, uint16_t holemask, diff --git a/fs/xfs/libxfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h index c5402bb4ce0c..77fffced8bac 100644 --- a/fs/xfs/libxfs/xfs_ialloc.h +++ b/fs/xfs/libxfs/xfs_ialloc.h @@ -176,6 +176,9 @@ int xfs_ialloc_has_inode_record(struct xfs_btree_cur *cur, xfs_agino_t low, xfs_agino_t high, bool *exists); int xfs_ialloc_count_inodes(struct xfs_btree_cur *cur, xfs_agino_t *count, xfs_agino_t *freecount); +int xfs_inobt_insert_rec(struct xfs_btree_cur *cur, uint16_t holemask, + uint8_t count, int32_t freecount, xfs_inofree_t free, + int *stat); int xfs_ialloc_cluster_alignment(struct xfs_mount *mp); void xfs_ialloc_agino_range(struct xfs_mount *mp, xfs_agnumber_t agno, diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index 367e9a0726e6..b04c55512159 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -296,7 +296,7 @@ xfs_inobt_verify( case cpu_to_be32(XFS_FIBT_MAGIC): break; default: - return NULL; + return __this_address; } /* level verification */ @@ -608,3 +608,12 @@ xfs_finobt_calc_reserves( *used += tree_len; return 0; } + +/* Calculate the inobt btree size for some records. */ +xfs_extlen_t +xfs_iallocbt_calc_size( + struct xfs_mount *mp, + unsigned long long len) +{ + return xfs_btree_calc_size(mp->m_inobt_mnr, len); +} diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h index aa81e2e63f3f..4acdd5458d59 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.h +++ b/fs/xfs/libxfs/xfs_ialloc_btree.h @@ -74,5 +74,7 @@ int xfs_inobt_rec_check_count(struct xfs_mount *, int xfs_finobt_calc_reserves(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_extlen_t *ask, xfs_extlen_t *used); +extern xfs_extlen_t xfs_iallocbt_calc_size(struct xfs_mount *mp, + unsigned long long len); #endif /* __XFS_IALLOC_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index ef68b1de006a..1201107eabc6 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -466,6 +466,8 @@ xfs_dinode_verify( return __this_address; if (di_size > XFS_DFORK_DSIZE(dip, mp)) return __this_address; + if (dip->di_nextents) + return __this_address; /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: @@ -484,12 +486,31 @@ xfs_dinode_verify( if (XFS_DFORK_Q(dip)) { switch (dip->di_aformat) { case XFS_DINODE_FMT_LOCAL: + if (dip->di_anextents) + return __this_address; + /* fall through */ case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: break; default: return __this_address; } + } else { + /* + * If there is no fork offset, this may be a freshly-made inode + * in a new disk cluster, in which case di_aformat is zeroed. + * Otherwise, such an inode must be in EXTENTS format; this goes + * for freed inodes as well. + */ + switch (dip->di_aformat) { + case 0: + case XFS_DINODE_FMT_EXTENTS: + break; + default: + return __this_address; + } + if (dip->di_anextents) + return __this_address; } /* only version 3 or greater inodes are extensively verified here */ diff --git a/fs/xfs/libxfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h index bb1b13a9b5f4..d4af2804b178 100644 --- a/fs/xfs/libxfs/xfs_quota_defs.h +++ b/fs/xfs/libxfs/xfs_quota_defs.h @@ -107,14 +107,12 @@ typedef uint16_t xfs_qwarncnt_t; * to a single function. None of these XFS_QMOPT_* flags are meant to have * persistent values (ie. their values can and will change between versions) */ -#define XFS_QMOPT_DQALLOC 0x0000002 /* alloc dquot ondisk if needed */ #define XFS_QMOPT_UQUOTA 0x0000004 /* user dquot requested */ #define XFS_QMOPT_PQUOTA 0x0000008 /* project dquot requested */ #define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */ #define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */ #define XFS_QMOPT_GQUOTA 0x0002000 /* group dquot requested */ #define XFS_QMOPT_ENOSPC 0x0004000 /* enospc instead of edquot (prj) */ -#define XFS_QMOPT_DQNEXT 0x0008000 /* return next dquot >= this ID */ /* * flags to xfs_trans_mod_dquot to indicate which field needs to be @@ -152,10 +150,11 @@ typedef uint16_t xfs_qwarncnt_t; #define XFS_QMOPT_RESBLK_MASK (XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS) extern xfs_failaddr_t xfs_dquot_verify(struct xfs_mount *mp, - struct xfs_disk_dquot *ddq, xfs_dqid_t id, uint type, - uint flags); + struct xfs_disk_dquot *ddq, xfs_dqid_t id, uint type); +extern xfs_failaddr_t xfs_dqblk_verify(struct xfs_mount *mp, + struct xfs_dqblk *dqb, xfs_dqid_t id, uint type); extern int xfs_calc_dquots_per_chunk(unsigned int nbblks); -extern int xfs_dquot_repair(struct xfs_mount *mp, struct xfs_disk_dquot *ddq, +extern int xfs_dqblk_repair(struct xfs_mount *mp, struct xfs_dqblk *dqb, xfs_dqid_t id, uint type); #endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/libxfs/xfs_refcount.c b/fs/xfs/libxfs/xfs_refcount.c index 560e28473024..418d53295893 100644 --- a/fs/xfs/libxfs/xfs_refcount.c +++ b/fs/xfs/libxfs/xfs_refcount.c @@ -88,8 +88,25 @@ xfs_refcount_lookup_ge( return xfs_btree_lookup(cur, XFS_LOOKUP_GE, stat); } +/* + * Look up the first record equal to [bno, len] in the btree + * given by cur. + */ +int +xfs_refcount_lookup_eq( + struct xfs_btree_cur *cur, + xfs_agblock_t bno, + int *stat) +{ + trace_xfs_refcount_lookup(cur->bc_mp, cur->bc_private.a.agno, bno, + XFS_LOOKUP_LE); + cur->bc_rec.rc.rc_startblock = bno; + cur->bc_rec.rc.rc_blockcount = 0; + return xfs_btree_lookup(cur, XFS_LOOKUP_EQ, stat); +} + /* Convert on-disk record to in-core format. */ -static inline void +void xfs_refcount_btrec_to_irec( union xfs_btree_rec *rec, struct xfs_refcount_irec *irec) @@ -149,7 +166,7 @@ xfs_refcount_update( * by [bno, len, refcount]. * This either works (return 0) or gets an EFSCORRUPTED error. */ -STATIC int +int xfs_refcount_insert( struct xfs_btree_cur *cur, struct xfs_refcount_irec *irec, @@ -162,7 +179,10 @@ xfs_refcount_insert( cur->bc_rec.rc.rc_blockcount = irec->rc_blockcount; cur->bc_rec.rc.rc_refcount = irec->rc_refcount; error = xfs_btree_insert(cur, i); + if (error) + goto out_error; XFS_WANT_CORRUPTED_GOTO(cur->bc_mp, *i == 1, out_error); + out_error: if (error) trace_xfs_refcount_insert_error(cur->bc_mp, diff --git a/fs/xfs/libxfs/xfs_refcount.h b/fs/xfs/libxfs/xfs_refcount.h index 2a731ac68fe4..a92ad9078bc1 100644 --- a/fs/xfs/libxfs/xfs_refcount.h +++ b/fs/xfs/libxfs/xfs_refcount.h @@ -24,6 +24,8 @@ extern int xfs_refcount_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno, int *stat); extern int xfs_refcount_lookup_ge(struct xfs_btree_cur *cur, xfs_agblock_t bno, int *stat); +extern int xfs_refcount_lookup_eq(struct xfs_btree_cur *cur, + xfs_agblock_t bno, int *stat); extern int xfs_refcount_get_rec(struct xfs_btree_cur *cur, struct xfs_refcount_irec *irec, int *stat); @@ -85,5 +87,10 @@ static inline xfs_fileoff_t xfs_refcount_max_unmap(int log_res) extern int xfs_refcount_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, bool *exists); +union xfs_btree_rec; +extern void xfs_refcount_btrec_to_irec(union xfs_btree_rec *rec, + struct xfs_refcount_irec *irec); +extern int xfs_refcount_insert(struct xfs_btree_cur *cur, + struct xfs_refcount_irec *irec, int *stat); #endif /* __XFS_REFCOUNT_H__ */ diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c index fba8d2718017..c0644f1be8a8 100644 --- a/fs/xfs/libxfs/xfs_rmap.c +++ b/fs/xfs/libxfs/xfs_rmap.c @@ -1374,6 +1374,8 @@ xfs_rmap_convert_shared( */ error = xfs_rmap_lookup_le_range(cur, bno, owner, offset, flags, &PREV, &i); + if (error) + goto done; XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done); ASSERT(PREV.rm_offset <= offset); @@ -2030,6 +2032,34 @@ out_error: return error; } +/* Insert a raw rmap into the rmapbt. */ +int +xfs_rmap_map_raw( + struct xfs_btree_cur *cur, + struct xfs_rmap_irec *rmap) +{ + struct xfs_owner_info oinfo; + + oinfo.oi_owner = rmap->rm_owner; + oinfo.oi_offset = rmap->rm_offset; + oinfo.oi_flags = 0; + if (rmap->rm_flags & XFS_RMAP_ATTR_FORK) + oinfo.oi_flags |= XFS_OWNER_INFO_ATTR_FORK; + if (rmap->rm_flags & XFS_RMAP_BMBT_BLOCK) + oinfo.oi_flags |= XFS_OWNER_INFO_BMBT_BLOCK; + + if (rmap->rm_flags || XFS_RMAP_NON_INODE_OWNER(rmap->rm_owner)) + return xfs_rmap_map(cur, rmap->rm_startblock, + rmap->rm_blockcount, + rmap->rm_flags & XFS_RMAP_UNWRITTEN, + &oinfo); + + return xfs_rmap_map_shared(cur, rmap->rm_startblock, + rmap->rm_blockcount, + rmap->rm_flags & XFS_RMAP_UNWRITTEN, + &oinfo); +} + struct xfs_rmap_query_range_info { xfs_rmap_query_range_fn fn; void *priv; @@ -2453,3 +2483,56 @@ xfs_rmap_record_exists( irec.rm_startblock + irec.rm_blockcount >= bno + len); return 0; } + +struct xfs_rmap_key_state { + uint64_t owner; + uint64_t offset; + unsigned int flags; + bool has_rmap; +}; + +/* For each rmap given, figure out if it doesn't match the key we want. */ +STATIC int +xfs_rmap_has_other_keys_helper( + struct xfs_btree_cur *cur, + struct xfs_rmap_irec *rec, + void *priv) +{ + struct xfs_rmap_key_state *rks = priv; + + if (rks->owner == rec->rm_owner && rks->offset == rec->rm_offset && + ((rks->flags & rec->rm_flags) & XFS_RMAP_KEY_FLAGS) == rks->flags) + return 0; + rks->has_rmap = true; + return XFS_BTREE_QUERY_RANGE_ABORT; +} + +/* + * Given an extent and some owner info, can we find records overlapping + * the extent whose owner info does not match the given owner? + */ +int +xfs_rmap_has_other_keys( + struct xfs_btree_cur *cur, + xfs_agblock_t bno, + xfs_extlen_t len, + struct xfs_owner_info *oinfo, + bool *has_rmap) +{ + struct xfs_rmap_irec low = {0}; + struct xfs_rmap_irec high; + struct xfs_rmap_key_state rks; + int error; + + xfs_owner_info_unpack(oinfo, &rks.owner, &rks.offset, &rks.flags); + rks.has_rmap = false; + + low.rm_startblock = bno; + memset(&high, 0xFF, sizeof(high)); + high.rm_startblock = bno + len - 1; + + error = xfs_rmap_query_range(cur, &low, &high, + xfs_rmap_has_other_keys_helper, &rks); + *has_rmap = rks.has_rmap; + return error; +} diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h index 380e53be98d5..43e506f67680 100644 --- a/fs/xfs/libxfs/xfs_rmap.h +++ b/fs/xfs/libxfs/xfs_rmap.h @@ -238,5 +238,9 @@ int xfs_rmap_has_record(struct xfs_btree_cur *cur, xfs_agblock_t bno, int xfs_rmap_record_exists(struct xfs_btree_cur *cur, xfs_agblock_t bno, xfs_extlen_t len, struct xfs_owner_info *oinfo, bool *has_rmap); +int xfs_rmap_has_other_keys(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len, struct xfs_owner_info *oinfo, + bool *has_rmap); +int xfs_rmap_map_raw(struct xfs_btree_cur *cur, struct xfs_rmap_irec *rmap); #endif /* __XFS_RMAP_H__ */ diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c index 106be2d0bb88..369eeb7a52ec 100644 --- a/fs/xfs/libxfs/xfs_rtbitmap.c +++ b/fs/xfs/libxfs/xfs_rtbitmap.c @@ -90,6 +90,9 @@ xfs_rtbuf_get( if (error) return error; + if (nmap == 0 || !xfs_bmap_is_real_extent(&map)) + return -EFSCORRUPTED; + ASSERT(map.br_startblock != NULLFSBLOCK); error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, XFS_FSB_TO_DADDR(mp, map.br_startblock), @@ -1033,14 +1036,17 @@ xfs_rtalloc_query_range( int is_free; int error = 0; - if (low_rec->ar_startblock > high_rec->ar_startblock) + if (low_rec->ar_startext > high_rec->ar_startext) return -EINVAL; - else if (low_rec->ar_startblock == high_rec->ar_startblock) + if (low_rec->ar_startext >= mp->m_sb.sb_rextents || + low_rec->ar_startext == high_rec->ar_startext) return 0; + if (high_rec->ar_startext >= mp->m_sb.sb_rextents) + high_rec->ar_startext = mp->m_sb.sb_rextents - 1; /* Iterate the bitmap, looking for discrepancies. */ - rtstart = low_rec->ar_startblock; - rem = high_rec->ar_startblock - rtstart; + rtstart = low_rec->ar_startext; + rem = high_rec->ar_startext - rtstart; while (rem) { /* Is the first block free? */ error = xfs_rtcheck_range(mp, tp, rtstart, 1, 1, &rtend, @@ -1050,13 +1056,13 @@ xfs_rtalloc_query_range( /* How long does the extent go for? */ error = xfs_rtfind_forw(mp, tp, rtstart, - high_rec->ar_startblock - 1, &rtend); + high_rec->ar_startext - 1, &rtend); if (error) break; if (is_free) { - rec.ar_startblock = rtstart; - rec.ar_blockcount = rtend - rtstart + 1; + rec.ar_startext = rtstart; + rec.ar_extcount = rtend - rtstart + 1; error = fn(tp, &rec, priv); if (error) @@ -1079,9 +1085,9 @@ xfs_rtalloc_query_all( { struct xfs_rtalloc_rec keys[2]; - keys[0].ar_startblock = 0; - keys[1].ar_startblock = tp->t_mountp->m_sb.sb_rblocks; - keys[0].ar_blockcount = keys[1].ar_blockcount = 0; + keys[0].ar_startext = 0; + keys[1].ar_startext = tp->t_mountp->m_sb.sb_rextents - 1; + keys[0].ar_extcount = keys[1].ar_extcount = 0; return xfs_rtalloc_query_range(tp, &keys[0], &keys[1], fn, priv); } diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index d9b94bd5f689..d485e14313c6 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -888,6 +888,109 @@ xfs_sync_sb( return xfs_trans_commit(tp); } +/* + * Update all the secondary superblocks to match the new state of the primary. + * Because we are completely overwriting all the existing fields in the + * secondary superblock buffers, there is no need to read them in from disk. + * Just get a new buffer, stamp it and write it. + * + * The sb buffers need to be cached here so that we serialise against other + * operations that access the secondary superblocks, but we don't want to keep + * them in memory once it is written so we mark it as a one-shot buffer. + */ +int +xfs_update_secondary_sbs( + struct xfs_mount *mp) +{ + xfs_agnumber_t agno; + int saved_error = 0; + int error = 0; + LIST_HEAD (buffer_list); + + /* update secondary superblocks. */ + for (agno = 1; agno < mp->m_sb.sb_agcount; agno++) { + struct xfs_buf *bp; + + bp = xfs_buf_get(mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_DADDR), + XFS_FSS_TO_BB(mp, 1), 0); + /* + * If we get an error reading or writing alternate superblocks, + * continue. xfs_repair chooses the "best" superblock based + * on most matches; if we break early, we'll leave more + * superblocks un-updated than updated, and xfs_repair may + * pick them over the properly-updated primary. + */ + if (!bp) { + xfs_warn(mp, + "error allocating secondary superblock for ag %d", + agno); + if (!saved_error) + saved_error = -ENOMEM; + continue; + } + + bp->b_ops = &xfs_sb_buf_ops; + xfs_buf_oneshot(bp); + xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); + xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb); + xfs_buf_delwri_queue(bp, &buffer_list); + xfs_buf_relse(bp); + + /* don't hold too many buffers at once */ + if (agno % 16) + continue; + + error = xfs_buf_delwri_submit(&buffer_list); + if (error) { + xfs_warn(mp, + "write error %d updating a secondary superblock near ag %d", + error, agno); + if (!saved_error) + saved_error = error; + continue; + } + } + error = xfs_buf_delwri_submit(&buffer_list); + if (error) { + xfs_warn(mp, + "write error %d updating a secondary superblock near ag %d", + error, agno); + } + + return saved_error ? saved_error : error; +} + +/* + * Same behavior as xfs_sync_sb, except that it is always synchronous and it + * also writes the superblock buffer to disk sector 0 immediately. + */ +int +xfs_sync_sb_buf( + struct xfs_mount *mp) +{ + struct xfs_trans *tp; + int error; + + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_sb, 0, 0, 0, &tp); + if (error) + return error; + + xfs_log_sb(tp); + xfs_trans_bhold(tp, mp->m_sb_bp); + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp); + if (error) + goto out; + /* + * write out the sb buffer to get the changes to disk + */ + error = xfs_bwrite(mp->m_sb_bp); +out: + xfs_buf_relse(mp->m_sb_bp); + return error; +} + int xfs_fs_geometry( struct xfs_sb *sbp, @@ -972,3 +1075,47 @@ xfs_fs_geometry( return 0; } + +/* Read a secondary superblock. */ +int +xfs_sb_read_secondary( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_agnumber_t agno, + struct xfs_buf **bpp) +{ + struct xfs_buf *bp; + int error; + + ASSERT(agno != 0 && agno != NULLAGNUMBER); + error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_BLOCK(mp)), + XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_sb_buf_ops); + if (error) + return error; + xfs_buf_set_ref(bp, XFS_SSB_REF); + *bpp = bp; + return 0; +} + +/* Get an uninitialised secondary superblock buffer. */ +int +xfs_sb_get_secondary( + struct xfs_mount *mp, + struct xfs_trans *tp, + xfs_agnumber_t agno, + struct xfs_buf **bpp) +{ + struct xfs_buf *bp; + + ASSERT(agno != 0 && agno != NULLAGNUMBER); + bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, + XFS_AG_DADDR(mp, agno, XFS_SB_BLOCK(mp)), + XFS_FSS_TO_BB(mp, 1), 0); + if (!bp) + return -ENOMEM; + bp->b_ops = &xfs_sb_buf_ops; + xfs_buf_oneshot(bp); + *bpp = bp; + return 0; +} diff --git a/fs/xfs/libxfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h index 63dcd2a1a657..244e0162c49e 100644 --- a/fs/xfs/libxfs/xfs_sb.h +++ b/fs/xfs/libxfs/xfs_sb.h @@ -18,6 +18,13 @@ #ifndef __XFS_SB_H__ #define __XFS_SB_H__ +struct xfs_mount; +struct xfs_sb; +struct xfs_dsb; +struct xfs_trans; +struct xfs_fsop_geom; +struct xfs_perag; + /* * perag get/put wrappers for ref counting */ @@ -29,13 +36,22 @@ extern int xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t); extern void xfs_log_sb(struct xfs_trans *tp); extern int xfs_sync_sb(struct xfs_mount *mp, bool wait); +extern int xfs_sync_sb_buf(struct xfs_mount *mp); extern void xfs_sb_mount_common(struct xfs_mount *mp, struct xfs_sb *sbp); extern void xfs_sb_from_disk(struct xfs_sb *to, struct xfs_dsb *from); extern void xfs_sb_to_disk(struct xfs_dsb *to, struct xfs_sb *from); extern void xfs_sb_quota_from_disk(struct xfs_sb *sbp); +extern int xfs_update_secondary_sbs(struct xfs_mount *mp); + #define XFS_FS_GEOM_MAX_STRUCT_VER (4) extern int xfs_fs_geometry(struct xfs_sb *sbp, struct xfs_fsop_geom *geo, int struct_version); +extern int xfs_sb_read_secondary(struct xfs_mount *mp, + struct xfs_trans *tp, xfs_agnumber_t agno, + struct xfs_buf **bpp); +extern int xfs_sb_get_secondary(struct xfs_mount *mp, + struct xfs_trans *tp, xfs_agnumber_t agno, + struct xfs_buf **bpp); #endif /* __XFS_SB_H__ */ diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h index d0b84da0cb1e..ae99c260adb1 100644 --- a/fs/xfs/libxfs/xfs_shared.h +++ b/fs/xfs/libxfs/xfs_shared.h @@ -57,21 +57,6 @@ extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops; extern const struct xfs_buf_ops xfs_symlink_buf_ops; extern const struct xfs_buf_ops xfs_rtbuf_ops; -/* - * This structure is used to track log items associated with - * a transaction. It points to the log item and keeps some - * flags to track the state of the log item. It also tracks - * the amount of space needed to log the item it describes - * once we get to commit processing (see xfs_trans_commit()). - */ -struct xfs_log_item_desc { - struct xfs_log_item *lid_item; - struct list_head lid_trans; - unsigned char lid_flags; -}; - -#define XFS_LID_DIRTY 0x1 - /* log size calculation functions */ int xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes); int xfs_log_calc_minimum_size(struct xfs_mount *); @@ -127,6 +112,7 @@ void xfs_log_get_max_trans_res(struct xfs_mount *mp, #define XFS_ATTR_BTREE_REF 1 #define XFS_DQUOT_REF 1 #define XFS_REFC_BTREE_REF 1 +#define XFS_SSB_REF 0 /* * Flags for xfs_trans_ichgtime(). diff --git a/fs/xfs/libxfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h index 3c560695c546..ea18449bd732 100644 --- a/fs/xfs/libxfs/xfs_types.h +++ b/fs/xfs/libxfs/xfs_types.h @@ -30,7 +30,7 @@ typedef int64_t xfs_fsize_t; /* bytes in a file */ typedef uint64_t xfs_ufsize_t; /* unsigned bytes in a file */ typedef int32_t xfs_suminfo_t; /* type of bitmap summary info */ -typedef int32_t xfs_rtword_t; /* word type for bitmap manipulations */ +typedef uint32_t xfs_rtword_t; /* word type for bitmap manipulations */ typedef int64_t xfs_lsn_t; /* log sequence number */ typedef int32_t xfs_tid_t; /* transaction identifier */ |