From 03dd4cb26d967f9588437b0fc9cc0e8353322bb7 Mon Sep 17 00:00:00 2001 From: André Fabian Silva Delgado Date: Fri, 25 Mar 2016 03:53:42 -0300 Subject: Linux-libre 4.5-gnu --- fs/xfs/kmem.h | 1 + fs/xfs/libxfs/xfs_alloc.c | 4 +- fs/xfs/libxfs/xfs_alloc.h | 1 + fs/xfs/libxfs/xfs_alloc_btree.c | 35 +- fs/xfs/libxfs/xfs_attr.c | 141 ++----- fs/xfs/libxfs/xfs_attr_leaf.c | 1 + fs/xfs/libxfs/xfs_attr_remote.c | 32 +- fs/xfs/libxfs/xfs_bit.c | 6 +- fs/xfs/libxfs/xfs_bmap.c | 43 ++- fs/xfs/libxfs/xfs_bmap.h | 2 +- fs/xfs/libxfs/xfs_bmap_btree.c | 1 + fs/xfs/libxfs/xfs_btree.c | 58 +++ fs/xfs/libxfs/xfs_btree.h | 3 + fs/xfs/libxfs/xfs_da_btree.c | 1 + fs/xfs/libxfs/xfs_dir2_block.c | 1 + fs/xfs/libxfs/xfs_dir2_data.c | 2 + fs/xfs/libxfs/xfs_dir2_leaf.c | 2 + fs/xfs/libxfs/xfs_dir2_node.c | 1 + fs/xfs/libxfs/xfs_dquot_buf.c | 37 +- fs/xfs/libxfs/xfs_format.h | 11 +- fs/xfs/libxfs/xfs_fs.h | 38 +- fs/xfs/libxfs/xfs_ialloc.c | 1 + fs/xfs/libxfs/xfs_ialloc_btree.c | 27 +- fs/xfs/libxfs/xfs_inode_buf.c | 4 + fs/xfs/libxfs/xfs_log_recover.h | 1 + fs/xfs/libxfs/xfs_quota_defs.h | 2 +- fs/xfs/libxfs/xfs_sb.c | 2 + fs/xfs/libxfs/xfs_shared.h | 1 + fs/xfs/libxfs/xfs_symlink_remote.c | 1 + fs/xfs/xfs_acl.c | 23 -- fs/xfs/xfs_acl.h | 4 - fs/xfs/xfs_aops.c | 8 +- fs/xfs/xfs_aops.h | 1 + fs/xfs/xfs_bmap_util.c | 46 +-- fs/xfs/xfs_buf.c | 10 +- fs/xfs/xfs_buf.h | 1 + fs/xfs/xfs_dquot.c | 13 +- fs/xfs/xfs_error.c | 4 +- fs/xfs/xfs_file.c | 38 +- fs/xfs/xfs_inode.c | 85 +++-- fs/xfs/xfs_ioctl.c | 92 +++-- fs/xfs/xfs_iomap.c | 21 +- fs/xfs/xfs_iops.c | 18 +- fs/xfs/xfs_log.c | 51 ++- fs/xfs/xfs_log_priv.h | 3 + fs/xfs/xfs_log_recover.c | 760 +++++++++++++++++++++++++++++-------- fs/xfs/xfs_pnfs.c | 4 +- fs/xfs/xfs_rtalloc.c | 3 +- fs/xfs/xfs_super.c | 6 +- fs/xfs/xfs_symlink.c | 12 +- fs/xfs/xfs_sysfs.c | 36 ++ fs/xfs/xfs_trace.h | 26 ++ fs/xfs/xfs_trans_dquot.c | 14 +- fs/xfs/xfs_xattr.c | 143 +++---- 54 files changed, 1153 insertions(+), 729 deletions(-) (limited to 'fs/xfs') diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h index cc6b768fc..d1c66e465 100644 --- a/fs/xfs/kmem.h +++ b/fs/xfs/kmem.h @@ -84,6 +84,7 @@ kmem_zalloc(size_t size, xfs_km_flags_t flags) #define KM_ZONE_HWALIGN SLAB_HWCACHE_ALIGN #define KM_ZONE_RECLAIM SLAB_RECLAIM_ACCOUNT #define KM_ZONE_SPREAD SLAB_MEM_SPREAD +#define KM_ZONE_ACCOUNT SLAB_ACCOUNT #define kmem_zone kmem_cache #define kmem_zone_t struct kmem_cache diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 3479294c1..a708e38b4 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -535,6 +535,7 @@ xfs_agfl_write_verify( } const struct xfs_buf_ops xfs_agfl_buf_ops = { + .name = "xfs_agfl", .verify_read = xfs_agfl_read_verify, .verify_write = xfs_agfl_write_verify, }; @@ -1926,7 +1927,7 @@ xfs_alloc_space_available( * Decide whether to use this allocation group for this allocation. * If so, fix up the btree freelist's size. */ -STATIC int /* error */ +int /* error */ xfs_alloc_fix_freelist( struct xfs_alloc_arg *args, /* allocation argument structure */ int flags) /* XFS_ALLOC_FLAG_... */ @@ -2339,6 +2340,7 @@ xfs_agf_write_verify( } const struct xfs_buf_ops xfs_agf_buf_ops = { + .name = "xfs_agf", .verify_read = xfs_agf_read_verify, .verify_write = xfs_agf_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 0ecde4d5c..135eb3d24 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -235,5 +235,6 @@ xfs_alloc_get_rec( 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_fix_freelist(struct xfs_alloc_arg *args, int flags); #endif /* __XFS_ALLOC_H__ */ diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c index 90de071dd..444626ddb 100644 --- a/fs/xfs/libxfs/xfs_alloc_btree.c +++ b/fs/xfs/libxfs/xfs_alloc_btree.c @@ -293,14 +293,7 @@ xfs_allocbt_verify( level = be16_to_cpu(block->bb_level); switch (block->bb_magic) { case cpu_to_be32(XFS_ABTB_CRC_MAGIC): - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return false; - if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) - return false; - if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn)) - return false; - if (pag && - be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno) + if (!xfs_btree_sblock_v5hdr_verify(bp)) return false; /* fall through */ case cpu_to_be32(XFS_ABTB_MAGIC): @@ -311,14 +304,7 @@ xfs_allocbt_verify( return false; break; case cpu_to_be32(XFS_ABTC_CRC_MAGIC): - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return false; - if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) - return false; - if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn)) - return false; - if (pag && - be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno) + if (!xfs_btree_sblock_v5hdr_verify(bp)) return false; /* fall through */ case cpu_to_be32(XFS_ABTC_MAGIC): @@ -332,21 +318,7 @@ xfs_allocbt_verify( return false; } - /* numrecs verification */ - if (be16_to_cpu(block->bb_numrecs) > mp->m_alloc_mxr[level != 0]) - return false; - - /* sibling pointer verification */ - if (!block->bb_u.s.bb_leftsib || - (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks && - block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK))) - return false; - if (!block->bb_u.s.bb_rightsib || - (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks && - block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK))) - return false; - - return true; + return xfs_btree_sblock_verify(bp, mp->m_alloc_mxr[level != 0]); } static void @@ -379,6 +351,7 @@ xfs_allocbt_write_verify( } const struct xfs_buf_ops xfs_allocbt_buf_ops = { + .name = "xfs_allocbt", .verify_read = xfs_allocbt_read_verify, .verify_write = xfs_allocbt_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index f949818fa..fa3b948ef 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -207,7 +207,7 @@ xfs_attr_set( struct xfs_trans_res tres; xfs_fsblock_t firstblock; int rsvd = (flags & ATTR_ROOT) != 0; - int error, err2, committed, local; + int error, err2, local; XFS_STATS_INC(mp, xs_attr_set); @@ -334,24 +334,14 @@ xfs_attr_set( */ xfs_bmap_init(args.flist, args.firstblock); error = xfs_attr_shortform_to_leaf(&args); - if (!error) { - error = xfs_bmap_finish(&args.trans, args.flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args.trans, args.flist, dp); if (error) { - ASSERT(committed); args.trans = NULL; xfs_bmap_cancel(&flist); goto out; } - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args.trans, dp, 0); - /* * Commit the leaf transformation. We'll need another (linked) * transaction to add the new attribute to the leaf. @@ -568,7 +558,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) { xfs_inode_t *dp; struct xfs_buf *bp; - int retval, error, committed, forkoff; + int retval, error, forkoff; trace_xfs_attr_leaf_addname(args); @@ -628,24 +618,14 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) */ xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_node(args); - if (!error) { - error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args->trans, args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); return error; } - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); - /* * Commit the current trans (including the inode) and start * a new one. @@ -729,25 +709,14 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ - if (!error) { + if (!error) error = xfs_bmap_finish(&args->trans, - args->flist, - &committed); - } + args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); return error; } - - /* - * bmap_finish() may have committed the last trans - * and started a new one. We need the inode to be - * in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); } /* @@ -775,7 +744,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args) { xfs_inode_t *dp; struct xfs_buf *bp; - int error, committed, forkoff; + int error, forkoff; trace_xfs_attr_leaf_removename(args); @@ -803,23 +772,13 @@ xfs_attr_leaf_removename(xfs_da_args_t *args) xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ - if (!error) { - error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args->trans, args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); return error; } - - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); } return 0; } @@ -877,7 +836,7 @@ xfs_attr_node_addname(xfs_da_args_t *args) xfs_da_state_blk_t *blk; xfs_inode_t *dp; xfs_mount_t *mp; - int committed, retval, error; + int retval, error; trace_xfs_attr_node_addname(args); @@ -938,26 +897,15 @@ restart: state = NULL; xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_node(args); - if (!error) { + if (!error) error = xfs_bmap_finish(&args->trans, - args->flist, - &committed); - } + args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); goto out; } - /* - * bmap_finish() may have committed the last trans - * and started a new one. We need the inode to be - * in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); - /* * Commit the node conversion and start the next * trans in the chain. @@ -977,23 +925,13 @@ restart: */ xfs_bmap_init(args->flist, args->firstblock); error = xfs_da3_split(state); - if (!error) { - error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args->trans, args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); goto out; } - - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); } else { /* * Addition succeeded, update Btree hashvals. @@ -1086,25 +1024,14 @@ restart: if (retval && (state->path.active > 1)) { xfs_bmap_init(args->flist, args->firstblock); error = xfs_da3_join(state); - if (!error) { + if (!error) error = xfs_bmap_finish(&args->trans, - args->flist, - &committed); - } + args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); goto out; } - - /* - * bmap_finish() may have committed the last trans - * and started a new one. We need the inode to be - * in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); } /* @@ -1146,7 +1073,7 @@ xfs_attr_node_removename(xfs_da_args_t *args) xfs_da_state_blk_t *blk; xfs_inode_t *dp; struct xfs_buf *bp; - int retval, error, committed, forkoff; + int retval, error, forkoff; trace_xfs_attr_node_removename(args); @@ -1220,24 +1147,13 @@ xfs_attr_node_removename(xfs_da_args_t *args) if (retval && (state->path.active > 1)) { xfs_bmap_init(args->flist, args->firstblock); error = xfs_da3_join(state); - if (!error) { - error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args->trans, args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); goto out; } - - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); - /* * Commit the Btree join operation and start a new trans. */ @@ -1265,25 +1181,14 @@ xfs_attr_node_removename(xfs_da_args_t *args) xfs_bmap_init(args->flist, args->firstblock); error = xfs_attr3_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ - if (!error) { + if (!error) error = xfs_bmap_finish(&args->trans, - args->flist, - &committed); - } + args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); goto out; } - - /* - * bmap_finish() may have committed the last trans - * and started a new one. We need the inode to be - * in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); } else xfs_trans_brelse(args->trans, bp); } diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index aa187f7ba..01a5ecfed 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -328,6 +328,7 @@ xfs_attr3_leaf_read_verify( } const struct xfs_buf_ops xfs_attr3_leaf_buf_ops = { + .name = "xfs_attr3_leaf", .verify_read = xfs_attr3_leaf_read_verify, .verify_write = xfs_attr3_leaf_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index 5ab95ffa4..a572532a5 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -201,6 +201,7 @@ xfs_attr3_rmt_write_verify( } const struct xfs_buf_ops xfs_attr3_rmt_buf_ops = { + .name = "xfs_attr3_rmt", .verify_read = xfs_attr3_rmt_read_verify, .verify_write = xfs_attr3_rmt_write_verify, }; @@ -447,8 +448,6 @@ xfs_attr_rmtval_set( * Roll through the "value", allocating blocks on disk as required. */ while (blkcnt > 0) { - int committed; - /* * Allocate a single extent, up to the size of the value. * @@ -466,24 +465,14 @@ xfs_attr_rmtval_set( error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno, blkcnt, XFS_BMAPI_ATTRFORK, args->firstblock, args->total, &map, &nmap, args->flist); - if (!error) { - error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + if (!error) + error = xfs_bmap_finish(&args->trans, args->flist, dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); return error; } - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, dp, 0); - ASSERT(nmap == 1); ASSERT((map.br_startblock != DELAYSTARTBLOCK) && (map.br_startblock != HOLESTARTBLOCK)); @@ -614,30 +603,19 @@ xfs_attr_rmtval_remove( blkcnt = args->rmtblkcnt; done = 0; while (!done) { - int committed; - xfs_bmap_init(args->flist, args->firstblock); error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt, XFS_BMAPI_ATTRFORK, 1, args->firstblock, args->flist, &done); - if (!error) { + if (!error) error = xfs_bmap_finish(&args->trans, args->flist, - &committed); - } + args->dp); if (error) { - ASSERT(committed); args->trans = NULL; xfs_bmap_cancel(args->flist); return error; } - /* - * bmap_finish() may have committed the last trans and started - * a new one. We need the inode to be in all transactions. - */ - if (committed) - xfs_trans_ijoin(args->trans, args->dp, 0); - /* * Close out trans and start the next one in the chain. */ diff --git a/fs/xfs/libxfs/xfs_bit.c b/fs/xfs/libxfs/xfs_bit.c index 0e8885a59..0a94cce5e 100644 --- a/fs/xfs/libxfs/xfs_bit.c +++ b/fs/xfs/libxfs/xfs_bit.c @@ -32,13 +32,13 @@ int xfs_bitmap_empty(uint *map, uint size) { uint i; - uint ret = 0; for (i = 0; i < size; i++) { - ret |= map[i]; + if (map[i] != 0) + return 0; } - return (ret == 0); + return 1; } /* diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 119c2422a..ef00156f4 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -325,9 +325,11 @@ xfs_check_block( /* * Check that the extents for the inode ip are in the right order in all - * btree leaves. + * btree leaves. THis becomes prohibitively expensive for large extent count + * files, so don't bother with inodes that have more than 10,000 extents in + * them. The btree record ordering checks will still be done, so for such large + * bmapbt constructs that is going to catch most corruptions. */ - STATIC void xfs_bmap_check_leaf_extents( xfs_btree_cur_t *cur, /* btree cursor or null */ @@ -352,6 +354,10 @@ xfs_bmap_check_leaf_extents( return; } + /* skip large extent count inodes */ + if (ip->i_d.di_nextents > 10000) + return; + bno = NULLFSBLOCK; mp = ip->i_mount; ifp = XFS_IFORK_PTR(ip, whichfork); @@ -1111,7 +1117,6 @@ xfs_bmap_add_attrfork( xfs_trans_t *tp; /* transaction pointer */ int blks; /* space reservation */ int version = 1; /* superblock attr version */ - int committed; /* xaction was committed */ int logflags; /* logging flags */ int error; /* error return value */ @@ -1214,7 +1219,7 @@ xfs_bmap_add_attrfork( xfs_log_sb(tp); } - error = xfs_bmap_finish(&tp, &flist, &committed); + error = xfs_bmap_finish(&tp, &flist, NULL); if (error) goto bmap_cancel; error = xfs_trans_commit(tp); @@ -1723,10 +1728,11 @@ xfs_bmap_add_extent_delay_real( xfs_filblks_t temp=0; /* value for da_new calculations */ xfs_filblks_t temp2=0;/* value for da_new calculations */ int tmp_rval; /* partial logging flags */ + int whichfork = XFS_DATA_FORK; struct xfs_mount *mp; - mp = bma->tp ? bma->tp->t_mountp : NULL; - ifp = XFS_IFORK_PTR(bma->ip, XFS_DATA_FORK); + mp = bma->ip->i_mount; + ifp = XFS_IFORK_PTR(bma->ip, whichfork); ASSERT(bma->idx >= 0); ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec)); @@ -1785,7 +1791,7 @@ xfs_bmap_add_extent_delay_real( * Don't set contiguous if the combined extent would be too large. * Also check for all-three-contiguous being too large. */ - if (bma->idx < bma->ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { + if (bma->idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) { state |= BMAP_RIGHT_VALID; xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx + 1), &RIGHT); @@ -2016,10 +2022,10 @@ xfs_bmap_add_extent_delay_real( XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done); } - if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) { + if (xfs_bmap_needs_btree(bma->ip, whichfork)) { error = xfs_bmap_extents_to_btree(bma->tp, bma->ip, bma->firstblock, bma->flist, - &bma->cur, 1, &tmp_rval, XFS_DATA_FORK); + &bma->cur, 1, &tmp_rval, whichfork); rval |= tmp_rval; if (error) goto done; @@ -2100,10 +2106,10 @@ xfs_bmap_add_extent_delay_real( XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done); } - if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) { + if (xfs_bmap_needs_btree(bma->ip, whichfork)) { error = xfs_bmap_extents_to_btree(bma->tp, bma->ip, bma->firstblock, bma->flist, &bma->cur, 1, - &tmp_rval, XFS_DATA_FORK); + &tmp_rval, whichfork); rval |= tmp_rval; if (error) goto done; @@ -2169,10 +2175,10 @@ xfs_bmap_add_extent_delay_real( XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done); } - if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) { + if (xfs_bmap_needs_btree(bma->ip, whichfork)) { error = xfs_bmap_extents_to_btree(bma->tp, bma->ip, bma->firstblock, bma->flist, &bma->cur, - 1, &tmp_rval, XFS_DATA_FORK); + 1, &tmp_rval, whichfork); rval |= tmp_rval; if (error) goto done; @@ -2215,13 +2221,13 @@ xfs_bmap_add_extent_delay_real( } /* convert to a btree if necessary */ - if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) { + if (xfs_bmap_needs_btree(bma->ip, whichfork)) { int tmp_logflags; /* partial log flag return val */ ASSERT(bma->cur == NULL); error = xfs_bmap_extents_to_btree(bma->tp, bma->ip, bma->firstblock, bma->flist, &bma->cur, - da_old > 0, &tmp_logflags, XFS_DATA_FORK); + da_old > 0, &tmp_logflags, whichfork); bma->logflags |= tmp_logflags; if (error) goto done; @@ -2242,7 +2248,7 @@ xfs_bmap_add_extent_delay_real( if (bma->cur) bma->cur->bc_private.b.allocated = 0; - xfs_bmap_check_leaf_extents(bma->cur, bma->ip, XFS_DATA_FORK); + xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork); done: bma->logflags |= rval; return error; @@ -2939,7 +2945,7 @@ xfs_bmap_add_extent_hole_real( int state; /* state bits, accessed thru macros */ struct xfs_mount *mp; - mp = bma->tp ? bma->tp->t_mountp : NULL; + mp = bma->ip->i_mount; ifp = XFS_IFORK_PTR(bma->ip, whichfork); ASSERT(bma->idx >= 0); @@ -5950,7 +5956,6 @@ xfs_bmap_split_extent( struct xfs_trans *tp; struct xfs_bmap_free free_list; xfs_fsblock_t firstfsb; - int committed; int error; tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT); @@ -5971,7 +5976,7 @@ xfs_bmap_split_extent( if (error) goto out; - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out; diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h index a160f8a5a..423a34e83 100644 --- a/fs/xfs/libxfs/xfs_bmap.h +++ b/fs/xfs/libxfs/xfs_bmap.h @@ -195,7 +195,7 @@ void xfs_bmap_add_free(xfs_fsblock_t bno, xfs_filblks_t len, struct xfs_bmap_free *flist, struct xfs_mount *mp); void xfs_bmap_cancel(struct xfs_bmap_free *flist); int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist, - int *committed); + struct xfs_inode *ip); 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); diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c index 6b0cf6546..1637c37bf 100644 --- a/fs/xfs/libxfs/xfs_bmap_btree.c +++ b/fs/xfs/libxfs/xfs_bmap_btree.c @@ -720,6 +720,7 @@ xfs_bmbt_write_verify( } const struct xfs_buf_ops xfs_bmbt_buf_ops = { + .name = "xfs_bmbt", .verify_read = xfs_bmbt_read_verify, .verify_write = xfs_bmbt_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index af1bbee55..a0eb18ce3 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -4080,3 +4080,61 @@ xfs_btree_change_owner( return 0; } + +/** + * xfs_btree_sblock_v5hdr_verify() -- verify the v5 fields of a short-format + * btree block + * + * @bp: buffer containing the btree block + * @max_recs: pointer to the m_*_mxr max records field in the xfs mount + * @pag_max_level: pointer to the per-ag max level field + */ +bool +xfs_btree_sblock_v5hdr_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); + struct xfs_perag *pag = bp->b_pag; + + if (!xfs_sb_version_hascrc(&mp->m_sb)) + return false; + if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) + return false; + if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn)) + return false; + if (pag && be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno) + return false; + return true; +} + +/** + * xfs_btree_sblock_verify() -- verify a short-format btree block + * + * @bp: buffer containing the btree block + * @max_recs: maximum records allowed in this btree node + */ +bool +xfs_btree_sblock_verify( + struct xfs_buf *bp, + unsigned int max_recs) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); + + /* numrecs verification */ + if (be16_to_cpu(block->bb_numrecs) > max_recs) + return false; + + /* sibling pointer verification */ + if (!block->bb_u.s.bb_leftsib || + (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks && + block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK))) + return false; + if (!block->bb_u.s.bb_rightsib || + (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks && + block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK))) + return false; + + return true; +} diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h index 992dec063..2e874be70 100644 --- a/fs/xfs/libxfs/xfs_btree.h +++ b/fs/xfs/libxfs/xfs_btree.h @@ -472,4 +472,7 @@ static inline int xfs_btree_get_level(struct xfs_btree_block *block) #define XFS_BTREE_TRACE_ARGR(c, r) #define XFS_BTREE_TRACE_CURSOR(c, t) +bool xfs_btree_sblock_v5hdr_verify(struct xfs_buf *bp); +bool xfs_btree_sblock_verify(struct xfs_buf *bp, unsigned int max_recs); + #endif /* __XFS_BTREE_H__ */ diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index e89a0f8f8..097bf7717 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -245,6 +245,7 @@ xfs_da3_node_read_verify( } const struct xfs_buf_ops xfs_da3_node_buf_ops = { + .name = "xfs_da3_node", .verify_read = xfs_da3_node_read_verify, .verify_write = xfs_da3_node_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c index 9c10e2b8c..aa17cb788 100644 --- a/fs/xfs/libxfs/xfs_dir2_block.c +++ b/fs/xfs/libxfs/xfs_dir2_block.c @@ -123,6 +123,7 @@ xfs_dir3_block_write_verify( } const struct xfs_buf_ops xfs_dir3_block_buf_ops = { + .name = "xfs_dir3_block", .verify_read = xfs_dir3_block_read_verify, .verify_write = xfs_dir3_block_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c index af71a84f3..725fc7841 100644 --- a/fs/xfs/libxfs/xfs_dir2_data.c +++ b/fs/xfs/libxfs/xfs_dir2_data.c @@ -305,11 +305,13 @@ xfs_dir3_data_write_verify( } const struct xfs_buf_ops xfs_dir3_data_buf_ops = { + .name = "xfs_dir3_data", .verify_read = xfs_dir3_data_read_verify, .verify_write = xfs_dir3_data_write_verify, }; static const struct xfs_buf_ops xfs_dir3_data_reada_buf_ops = { + .name = "xfs_dir3_data_reada", .verify_read = xfs_dir3_data_reada_verify, .verify_write = xfs_dir3_data_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c index 3923e1f94..b887fb2a2 100644 --- a/fs/xfs/libxfs/xfs_dir2_leaf.c +++ b/fs/xfs/libxfs/xfs_dir2_leaf.c @@ -245,11 +245,13 @@ xfs_dir3_leafn_write_verify( } const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops = { + .name = "xfs_dir3_leaf1", .verify_read = xfs_dir3_leaf1_read_verify, .verify_write = xfs_dir3_leaf1_write_verify, }; const struct xfs_buf_ops xfs_dir3_leafn_buf_ops = { + .name = "xfs_dir3_leafn", .verify_read = xfs_dir3_leafn_read_verify, .verify_write = xfs_dir3_leafn_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index 70b0cb2fd..63ee03db7 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -150,6 +150,7 @@ xfs_dir3_free_write_verify( } const struct xfs_buf_ops xfs_dir3_free_buf_ops = { + .name = "xfs_dir3_free", .verify_read = xfs_dir3_free_read_verify, .verify_write = xfs_dir3_free_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c index 5331b7f04..3cc3cf767 100644 --- a/fs/xfs/libxfs/xfs_dquot_buf.c +++ b/fs/xfs/libxfs/xfs_dquot_buf.c @@ -54,7 +54,7 @@ xfs_dqcheck( xfs_dqid_t id, uint type, /* used only when IO_dorepair is true */ uint flags, - char *str) + const char *str) { xfs_dqblk_t *d = (xfs_dqblk_t *)ddq; int errs = 0; @@ -207,7 +207,8 @@ xfs_dquot_buf_verify_crc( STATIC bool xfs_dquot_buf_verify( struct xfs_mount *mp, - struct xfs_buf *bp) + struct xfs_buf *bp, + int warn) { struct xfs_dqblk *d = (struct xfs_dqblk *)bp->b_addr; xfs_dqid_t id = 0; @@ -240,8 +241,7 @@ xfs_dquot_buf_verify( if (i == 0) id = be32_to_cpu(ddq->d_id); - error = xfs_dqcheck(mp, ddq, id + i, 0, XFS_QMOPT_DOWARN, - "xfs_dquot_buf_verify"); + error = xfs_dqcheck(mp, ddq, id + i, 0, warn, __func__); if (error) return false; } @@ -256,13 +256,32 @@ xfs_dquot_buf_read_verify( if (!xfs_dquot_buf_verify_crc(mp, bp)) xfs_buf_ioerror(bp, -EFSBADCRC); - else if (!xfs_dquot_buf_verify(mp, bp)) + else if (!xfs_dquot_buf_verify(mp, bp, XFS_QMOPT_DOWARN)) xfs_buf_ioerror(bp, -EFSCORRUPTED); if (bp->b_error) xfs_verifier_error(bp); } +/* + * readahead errors are silent and simply leave the buffer as !done so a real + * read will then be run with the xfs_dquot_buf_ops verifier. See + * xfs_inode_buf_verify() for why we use EIO and ~XBF_DONE here rather than + * reporting the failure. + */ +static void +xfs_dquot_buf_readahead_verify( + struct xfs_buf *bp) +{ + struct xfs_mount *mp = bp->b_target->bt_mount; + + if (!xfs_dquot_buf_verify_crc(mp, bp) || + !xfs_dquot_buf_verify(mp, bp, 0)) { + xfs_buf_ioerror(bp, -EIO); + bp->b_flags &= ~XBF_DONE; + } +} + /* * we don't calculate the CRC here as that is done when the dquot is flushed to * the buffer after the update is done. This ensures that the dquot in the @@ -274,7 +293,7 @@ xfs_dquot_buf_write_verify( { struct xfs_mount *mp = bp->b_target->bt_mount; - if (!xfs_dquot_buf_verify(mp, bp)) { + if (!xfs_dquot_buf_verify(mp, bp, XFS_QMOPT_DOWARN)) { xfs_buf_ioerror(bp, -EFSCORRUPTED); xfs_verifier_error(bp); return; @@ -282,7 +301,13 @@ xfs_dquot_buf_write_verify( } const struct xfs_buf_ops xfs_dquot_buf_ops = { + .name = "xfs_dquot", .verify_read = xfs_dquot_buf_read_verify, .verify_write = xfs_dquot_buf_write_verify, }; +const struct xfs_buf_ops xfs_dquot_buf_ra_ops = { + .name = "xfs_dquot_ra", + .verify_read = xfs_dquot_buf_readahead_verify, + .verify_write = xfs_dquot_buf_write_verify, +}; diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h index e2536bb1c..dc97eb21a 100644 --- a/fs/xfs/libxfs/xfs_format.h +++ b/fs/xfs/libxfs/xfs_format.h @@ -984,8 +984,6 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev) /* * Values for di_flags - * There should be a one-to-one correspondence between these flags and the - * XFS_XFLAG_s. */ #define XFS_DIFLAG_REALTIME_BIT 0 /* file's blocks come from rt area */ #define XFS_DIFLAG_PREALLOC_BIT 1 /* file space has been preallocated */ @@ -1025,6 +1023,15 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev) XFS_DIFLAG_PROJINHERIT | XFS_DIFLAG_NOSYMLINKS | XFS_DIFLAG_EXTSIZE | \ XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG | XFS_DIFLAG_FILESTREAM) +/* + * Values for di_flags2 These start by being exposed to userspace in the upper + * 16 bits of the XFS_XFLAG_s range. + */ +#define XFS_DIFLAG2_DAX_BIT 0 /* use DAX for this inode */ +#define XFS_DIFLAG2_DAX (1 << XFS_DIFLAG2_DAX_BIT) + +#define XFS_DIFLAG2_ANY (XFS_DIFLAG2_DAX) + /* * Inode number format: * low inopblog bits - offset in block diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index b2b73a998..fffe3d01b 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -35,40 +35,6 @@ struct dioattr { }; #endif -/* - * Structure for XFS_IOC_FSGETXATTR[A] and XFS_IOC_FSSETXATTR. - */ -#ifndef HAVE_FSXATTR -struct fsxattr { - __u32 fsx_xflags; /* xflags field value (get/set) */ - __u32 fsx_extsize; /* extsize field value (get/set)*/ - __u32 fsx_nextents; /* nextents field value (get) */ - __u32 fsx_projid; /* project identifier (get/set) */ - unsigned char fsx_pad[12]; -}; -#endif - -/* - * Flags for the bs_xflags/fsx_xflags field - * There should be a one-to-one correspondence between these flags and the - * XFS_DIFLAG_s. - */ -#define XFS_XFLAG_REALTIME 0x00000001 /* data in realtime volume */ -#define XFS_XFLAG_PREALLOC 0x00000002 /* preallocated file extents */ -#define XFS_XFLAG_IMMUTABLE 0x00000008 /* file cannot be modified */ -#define XFS_XFLAG_APPEND 0x00000010 /* all writes append */ -#define XFS_XFLAG_SYNC 0x00000020 /* all writes synchronous */ -#define XFS_XFLAG_NOATIME 0x00000040 /* do not update access time */ -#define XFS_XFLAG_NODUMP 0x00000080 /* do not include in backups */ -#define XFS_XFLAG_RTINHERIT 0x00000100 /* create with rt bit set */ -#define XFS_XFLAG_PROJINHERIT 0x00000200 /* create with parents projid */ -#define XFS_XFLAG_NOSYMLINKS 0x00000400 /* disallow symlink creation */ -#define XFS_XFLAG_EXTSIZE 0x00000800 /* extent size allocator hint */ -#define XFS_XFLAG_EXTSZINHERIT 0x00001000 /* inherit inode extent size */ -#define XFS_XFLAG_NODEFRAG 0x00002000 /* do not defragment */ -#define XFS_XFLAG_FILESTREAM 0x00004000 /* use filestream allocator */ -#define XFS_XFLAG_HASATTR 0x80000000 /* no DIFLAG for this */ - /* * Structure for XFS_IOC_GETBMAP. * On input, fill in bmv_offset and bmv_length of the first structure @@ -514,8 +480,8 @@ typedef struct xfs_swapext #define XFS_IOC_ALLOCSP _IOW ('X', 10, struct xfs_flock64) #define XFS_IOC_FREESP _IOW ('X', 11, struct xfs_flock64) #define XFS_IOC_DIOINFO _IOR ('X', 30, struct dioattr) -#define XFS_IOC_FSGETXATTR _IOR ('X', 31, struct fsxattr) -#define XFS_IOC_FSSETXATTR _IOW ('X', 32, struct fsxattr) +#define XFS_IOC_FSGETXATTR FS_IOC_FSGETXATTR +#define XFS_IOC_FSSETXATTR FS_IOC_FSSETXATTR #define XFS_IOC_ALLOCSP64 _IOW ('X', 36, struct xfs_flock64) #define XFS_IOC_FREESP64 _IOW ('X', 37, struct xfs_flock64) #define XFS_IOC_GETBMAP _IOWR('X', 38, struct getbmap) diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index 70c1db99f..66d702e6b 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -2572,6 +2572,7 @@ xfs_agi_write_verify( } const struct xfs_buf_ops xfs_agi_buf_ops = { + .name = "xfs_agi", .verify_read = xfs_agi_read_verify, .verify_write = xfs_agi_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c index f39b285be..c679f3c05 100644 --- a/fs/xfs/libxfs/xfs_ialloc_btree.c +++ b/fs/xfs/libxfs/xfs_ialloc_btree.c @@ -221,7 +221,6 @@ xfs_inobt_verify( { struct xfs_mount *mp = bp->b_target->bt_mount; struct xfs_btree_block *block = XFS_BUF_TO_BLOCK(bp); - struct xfs_perag *pag = bp->b_pag; unsigned int level; /* @@ -237,14 +236,7 @@ xfs_inobt_verify( switch (block->bb_magic) { case cpu_to_be32(XFS_IBT_CRC_MAGIC): case cpu_to_be32(XFS_FIBT_CRC_MAGIC): - if (!xfs_sb_version_hascrc(&mp->m_sb)) - return false; - if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid)) - return false; - if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn)) - return false; - if (pag && - be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno) + if (!xfs_btree_sblock_v5hdr_verify(bp)) return false; /* fall through */ case cpu_to_be32(XFS_IBT_MAGIC): @@ -254,24 +246,12 @@ xfs_inobt_verify( return 0; } - /* numrecs and level verification */ + /* level verification */ level = be16_to_cpu(block->bb_level); if (level >= mp->m_in_maxlevels) return false; - if (be16_to_cpu(block->bb_numrecs) > mp->m_inobt_mxr[level != 0]) - return false; - - /* sibling pointer verification */ - if (!block->bb_u.s.bb_leftsib || - (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks && - block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK))) - return false; - if (!block->bb_u.s.bb_rightsib || - (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks && - block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK))) - return false; - return true; + return xfs_btree_sblock_verify(bp, mp->m_inobt_mxr[level != 0]); } static void @@ -304,6 +284,7 @@ xfs_inobt_write_verify( } const struct xfs_buf_ops xfs_inobt_buf_ops = { + .name = "xfs_inobt", .verify_read = xfs_inobt_read_verify, .verify_write = xfs_inobt_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index 65485cfc4..1aabfda66 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -68,6 +68,8 @@ xfs_inobp_check( * recovery and we don't get unnecssary panics on debug kernels. We use EIO here * because all we want to do is say readahead failed; there is no-one to report * the error to, so this will distinguish it from a non-ra verifier failure. + * Changes to this readahead error behavour also need to be reflected in + * xfs_dquot_buf_readahead_verify(). */ static void xfs_inode_buf_verify( @@ -134,11 +136,13 @@ xfs_inode_buf_write_verify( } const struct xfs_buf_ops xfs_inode_buf_ops = { + .name = "xfs_inode", .verify_read = xfs_inode_buf_read_verify, .verify_write = xfs_inode_buf_write_verify, }; const struct xfs_buf_ops xfs_inode_buf_ra_ops = { + .name = "xxfs_inode_ra", .verify_read = xfs_inode_buf_readahead_verify, .verify_write = xfs_inode_buf_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h index 1c55ccbb3..8e385f91d 100644 --- a/fs/xfs/libxfs/xfs_log_recover.h +++ b/fs/xfs/libxfs/xfs_log_recover.h @@ -60,6 +60,7 @@ typedef struct xlog_recover { */ #define XLOG_BC_TABLE_SIZE 64 +#define XLOG_RECOVER_CRCPASS 0 #define XLOG_RECOVER_PASS1 1 #define XLOG_RECOVER_PASS2 2 diff --git a/fs/xfs/libxfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h index 1b0a08379..f51078f1e 100644 --- a/fs/xfs/libxfs/xfs_quota_defs.h +++ b/fs/xfs/libxfs/xfs_quota_defs.h @@ -153,7 +153,7 @@ typedef __uint16_t xfs_qwarncnt_t; #define XFS_QMOPT_RESBLK_MASK (XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS) extern int xfs_dqcheck(struct xfs_mount *mp, xfs_disk_dquot_t *ddq, - xfs_dqid_t id, uint type, uint flags, char *str); + xfs_dqid_t id, uint type, uint flags, const char *str); extern int xfs_calc_dquots_per_chunk(unsigned int nbblks); #endif /* __XFS_QUOTA_H__ */ diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index a0b071d88..8a53eaa34 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -679,11 +679,13 @@ xfs_sb_write_verify( } const struct xfs_buf_ops xfs_sb_buf_ops = { + .name = "xfs_sb", .verify_read = xfs_sb_read_verify, .verify_write = xfs_sb_write_verify, }; const struct xfs_buf_ops xfs_sb_quiet_buf_ops = { + .name = "xfs_sb_quiet", .verify_read = xfs_sb_quiet_read_verify, .verify_write = xfs_sb_write_verify, }; diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h index 5be529707..15c3ceb84 100644 --- a/fs/xfs/libxfs/xfs_shared.h +++ b/fs/xfs/libxfs/xfs_shared.h @@ -49,6 +49,7 @@ extern const struct xfs_buf_ops xfs_inobt_buf_ops; extern const struct xfs_buf_ops xfs_inode_buf_ops; extern const struct xfs_buf_ops xfs_inode_buf_ra_ops; extern const struct xfs_buf_ops xfs_dquot_buf_ops; +extern const struct xfs_buf_ops xfs_dquot_buf_ra_ops; extern const struct xfs_buf_ops xfs_sb_buf_ops; extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops; extern const struct xfs_buf_ops xfs_symlink_buf_ops; diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c index cb6fd20a4..2e2c6716b 100644 --- a/fs/xfs/libxfs/xfs_symlink_remote.c +++ b/fs/xfs/libxfs/xfs_symlink_remote.c @@ -168,6 +168,7 @@ xfs_symlink_write_verify( } const struct xfs_buf_ops xfs_symlink_buf_ops = { + .name = "xfs_symlink", .verify_read = xfs_symlink_read_verify, .verify_write = xfs_symlink_write_verify, }; diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 6bb470fbb..2d5df1f23 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -252,29 +252,6 @@ xfs_set_mode(struct inode *inode, umode_t mode) return error; } -static int -xfs_acl_exists(struct inode *inode, unsigned char *name) -{ - int len = XFS_ACL_MAX_SIZE(XFS_M(inode->i_sb)); - - return (xfs_attr_get(XFS_I(inode), name, NULL, &len, - ATTR_ROOT|ATTR_KERNOVAL) == 0); -} - -int -posix_acl_access_exists(struct inode *inode) -{ - return xfs_acl_exists(inode, SGI_ACL_FILE); -} - -int -posix_acl_default_exists(struct inode *inode) -{ - if (!S_ISDIR(inode->i_mode)) - return 0; - return xfs_acl_exists(inode, SGI_ACL_DEFAULT); -} - int xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) { diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 52f8255d6..286fa8921 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -24,16 +24,12 @@ struct posix_acl; #ifdef CONFIG_XFS_POSIX_ACL extern struct posix_acl *xfs_get_acl(struct inode *inode, int type); extern int xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); -extern int posix_acl_access_exists(struct inode *inode); -extern int posix_acl_default_exists(struct inode *inode); #else static inline struct posix_acl *xfs_get_acl(struct inode *inode, int type) { return NULL; } # define xfs_set_acl NULL -# define posix_acl_access_exists(inode) 0 -# define posix_acl_default_exists(inode) 0 #endif /* CONFIG_XFS_POSIX_ACL */ extern void xfs_forget_acl(struct inode *inode, const char *name, int xflags); diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 29e7e5dd5..a9ebabfe7 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -55,7 +55,7 @@ xfs_count_page_state( } while ((bh = bh->b_this_page) != head); } -STATIC struct block_device * +struct block_device * xfs_find_bdev_for_inode( struct inode *inode) { @@ -1208,6 +1208,10 @@ xfs_vm_writepages( struct writeback_control *wbc) { xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); + if (dax_mapping(mapping)) + return dax_writeback_mapping_range(mapping, + xfs_find_bdev_for_inode(mapping->host), wbc); + return generic_writepages(mapping, wbc); } @@ -1917,6 +1921,7 @@ xfs_vm_readpage( struct file *unused, struct page *page) { + trace_xfs_vm_readpage(page->mapping->host, 1); return mpage_readpage(page, xfs_get_blocks); } @@ -1927,6 +1932,7 @@ xfs_vm_readpages( struct list_head *pages, unsigned nr_pages) { + trace_xfs_vm_readpages(mapping->host, nr_pages); return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks); } diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h index f6ffc9ae5..a4343c63f 100644 --- a/fs/xfs/xfs_aops.h +++ b/fs/xfs/xfs_aops.h @@ -62,5 +62,6 @@ int xfs_get_blocks_dax_fault(struct inode *inode, sector_t offset, struct buffer_head *map_bh, int create); extern void xfs_count_page_state(struct page *, int *, int *); +extern struct block_device *xfs_find_bdev_for_inode(struct inode *); #endif /* __XFS_AOPS_H__ */ diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index dbae6490a..6c876012b 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -75,7 +75,8 @@ xfs_zero_extent( ssize_t size = XFS_FSB_TO_B(mp, count_fsb); if (IS_DAX(VFS_I(ip))) - return dax_clear_blocks(VFS_I(ip), block, size); + return dax_clear_sectors(xfs_find_bdev_for_inode(VFS_I(ip)), + sector, size); /* * let the block layer decide on the fastest method of @@ -91,32 +92,32 @@ xfs_zero_extent( * last due to locking considerations. We never free any extents in * the first transaction. * - * Return 1 if the given transaction was committed and a new one - * started, and 0 otherwise in the committed parameter. + * If an inode *ip is provided, rejoin it to the transaction if + * the transaction was committed. */ int /* error */ xfs_bmap_finish( struct xfs_trans **tp, /* transaction pointer addr */ struct xfs_bmap_free *flist, /* i/o: list extents to free */ - int *committed)/* xact committed or not */ + struct xfs_inode *ip) { struct xfs_efd_log_item *efd; /* extent free data */ struct xfs_efi_log_item *efi; /* extent free intention */ int error; /* error return value */ + int committed;/* xact committed or not */ struct xfs_bmap_free_item *free; /* free extent item */ struct xfs_bmap_free_item *next; /* next item on free list */ ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES); - if (flist->xbf_count == 0) { - *committed = 0; + if (flist->xbf_count == 0) return 0; - } + efi = xfs_trans_get_efi(*tp, flist->xbf_count); for (free = flist->xbf_first; free; free = free->xbfi_next) xfs_trans_log_efi_extent(*tp, efi, free->xbfi_startblock, free->xbfi_blockcount); - error = __xfs_trans_roll(tp, NULL, committed); + error = __xfs_trans_roll(tp, ip, &committed); if (error) { /* * If the transaction was committed, drop the EFD reference @@ -128,16 +129,13 @@ xfs_bmap_finish( * transaction so we should return committed=1 even though we're * returning an error. */ - if (*committed) { + if (committed) { xfs_efi_release(efi); xfs_force_shutdown((*tp)->t_mountp, (error == -EFSCORRUPTED) ? SHUTDOWN_CORRUPT_INCORE : SHUTDOWN_META_IO_ERROR); - } else { - *committed = 1; } - return error; } @@ -969,7 +967,6 @@ xfs_alloc_file_space( xfs_bmbt_irec_t imaps[1], *imapp; xfs_bmap_free_t free_list; uint qblocks, resblks, resrtextents; - int committed; int error; trace_xfs_alloc_file_space(ip); @@ -1064,23 +1061,20 @@ xfs_alloc_file_space( error = xfs_bmapi_write(tp, ip, startoffset_fsb, allocatesize_fsb, alloc_type, &firstfsb, resblks, imapp, &nimaps, &free_list); - if (error) { + if (error) goto error0; - } /* * Complete the transaction */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { + error = xfs_bmap_finish(&tp, &free_list, NULL); + if (error) goto error0; - } error = xfs_trans_commit(tp); xfs_iunlock(ip, XFS_ILOCK_EXCL); - if (error) { + if (error) break; - } allocated_fsb = imapp->br_blockcount; @@ -1206,7 +1200,6 @@ xfs_free_file_space( xfs_off_t offset, xfs_off_t len) { - int committed; int done; xfs_fileoff_t endoffset_fsb; int error; @@ -1346,17 +1339,15 @@ xfs_free_file_space( error = xfs_bunmapi(tp, ip, startoffset_fsb, endoffset_fsb - startoffset_fsb, 0, 2, &firstfsb, &free_list, &done); - if (error) { + if (error) goto error0; - } /* * complete the transaction */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (error) { + error = xfs_bmap_finish(&tp, &free_list, NULL); + if (error) goto error0; - } error = xfs_trans_commit(tp); xfs_iunlock(ip, XFS_ILOCK_EXCL); @@ -1434,7 +1425,6 @@ xfs_shift_file_space( int error; struct xfs_bmap_free free_list; xfs_fsblock_t first_block; - int committed; xfs_fileoff_t stop_fsb; xfs_fileoff_t next_fsb; xfs_fileoff_t shift_fsb; @@ -1526,7 +1516,7 @@ xfs_shift_file_space( if (error) goto out_bmap_cancel; - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out_bmap_cancel; diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index 39090fc56..435c7de42 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1052,7 +1052,7 @@ xfs_buf_ioend_work( xfs_buf_ioend(bp); } -void +static void xfs_buf_ioend_async( struct xfs_buf *bp) { @@ -1649,13 +1649,9 @@ xfs_setsize_buftarg( btp->bt_meta_sectormask = sectorsize - 1; if (set_blocksize(btp->bt_bdev, sectorsize)) { - char name[BDEVNAME_SIZE]; - - bdevname(btp->bt_bdev, name); - xfs_warn(btp->bt_mount, - "Cannot set_blocksize to %u on device %s", - sectorsize, name); + "Cannot set_blocksize to %u on device %pg", + sectorsize, btp->bt_bdev); return -EINVAL; } diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index c79b717d9..c75721acd 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -132,6 +132,7 @@ struct xfs_buf_map { struct xfs_buf_map (map) = { .bm_bn = (blkno), .bm_len = (numblk) }; struct xfs_buf_ops { + char *name; void (*verify_read)(struct xfs_buf *); void (*verify_write)(struct xfs_buf *); }; diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 7ac6c5c58..9c44d38dc 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -306,7 +306,7 @@ xfs_qm_dqalloc( xfs_fsblock_t firstblock; xfs_bmap_free_t flist; xfs_bmbt_irec_t map; - int nmaps, error, committed; + int nmaps, error; xfs_buf_t *bp; xfs_trans_t *tp = *tpp; @@ -379,11 +379,12 @@ xfs_qm_dqalloc( xfs_trans_bhold(tp, bp); - if ((error = xfs_bmap_finish(tpp, &flist, &committed))) { + error = xfs_bmap_finish(tpp, &flist, NULL); + if (error) goto error1; - } - if (committed) { + /* Transaction was committed? */ + if (*tpp != tp) { tp = *tpp; xfs_trans_bjoin(tp, bp); } else { @@ -393,9 +394,9 @@ xfs_qm_dqalloc( *O_bpp = bp; return 0; - error1: +error1: xfs_bmap_cancel(&flist); - error0: +error0: xfs_iunlock(quotip, XFS_ILOCK_EXCL); return error; diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index 74d0e5966..88693a98f 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -164,9 +164,9 @@ xfs_verifier_error( { struct xfs_mount *mp = bp->b_target->bt_mount; - xfs_alert(mp, "Metadata %s detected at %pF, block 0x%llx", + xfs_alert(mp, "Metadata %s detected at %pF, %s block 0x%llx", bp->b_error == -EFSBADCRC ? "CRC error" : "corruption", - __return_address, bp->b_bn); + __return_address, bp->b_ops->name, bp->b_bn); xfs_alert(mp, "Unmount and run xfs_repair"); diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index f5392ab2d..52883ac3c 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -55,7 +55,7 @@ xfs_rw_ilock( int type) { if (type & XFS_IOLOCK_EXCL) - mutex_lock(&VFS_I(ip)->i_mutex); + inode_lock(VFS_I(ip)); xfs_ilock(ip, type); } @@ -66,7 +66,7 @@ xfs_rw_iunlock( { xfs_iunlock(ip, type); if (type & XFS_IOLOCK_EXCL) - mutex_unlock(&VFS_I(ip)->i_mutex); + inode_unlock(VFS_I(ip)); } static inline void @@ -76,7 +76,7 @@ xfs_rw_ilock_demote( { xfs_ilock_demote(ip, type); if (type & XFS_IOLOCK_EXCL) - mutex_unlock(&VFS_I(ip)->i_mutex); + inode_unlock(VFS_I(ip)); } /* @@ -402,19 +402,26 @@ xfs_file_splice_read( if (XFS_FORCED_SHUTDOWN(ip->i_mount)) return -EIO; - xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); - trace_xfs_file_splice_read(ip, count, *ppos, ioflags); - /* for dax, we need to avoid the page cache */ - if (IS_DAX(VFS_I(ip))) - ret = default_file_splice_read(infilp, ppos, pipe, count, flags); - else - ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); - if (ret > 0) - XFS_STATS_ADD(ip->i_mount, xs_read_bytes, ret); + /* + * DAX inodes cannot ues the page cache for splice, so we have to push + * them through the VFS IO path. This means it goes through + * ->read_iter, which for us takes the XFS_IOLOCK_SHARED. Hence we + * cannot lock the splice operation at this level for DAX inodes. + */ + if (IS_DAX(VFS_I(ip))) { + ret = default_file_splice_read(infilp, ppos, pipe, count, + flags); + goto out; + } + xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); + ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED); +out: + if (ret > 0) + XFS_STATS_ADD(ip->i_mount, xs_read_bytes, ret); return ret; } @@ -1603,9 +1610,8 @@ xfs_filemap_pmd_fault( /* * pfn_mkwrite was originally inteneded to ensure we capture time stamp * updates on write faults. In reality, it's need to serialise against - * truncate similar to page_mkwrite. Hence we open-code dax_pfn_mkwrite() - * here and cycle the XFS_MMAPLOCK_SHARED to ensure we serialise the fault - * barrier in place. + * truncate similar to page_mkwrite. Hence we cycle the XFS_MMAPLOCK_SHARED + * to ensure we serialise the fault barrier in place. */ static int xfs_filemap_pfn_mkwrite( @@ -1628,6 +1634,8 @@ xfs_filemap_pfn_mkwrite( size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT; if (vmf->pgoff >= size) ret = VM_FAULT_SIGBUS; + else if (IS_DAX(inode)) + ret = dax_pfn_mkwrite(vma, vmf); xfs_iunlock(ip, XFS_MMAPLOCK_SHARED); sb_end_pagefault(inode->i_sb); return ret; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index 8ee393996..ceba1a83c 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -610,60 +610,69 @@ __xfs_iflock( STATIC uint _xfs_dic2xflags( - __uint16_t di_flags) + __uint16_t di_flags, + uint64_t di_flags2, + bool has_attr) { uint flags = 0; if (di_flags & XFS_DIFLAG_ANY) { if (di_flags & XFS_DIFLAG_REALTIME) - flags |= XFS_XFLAG_REALTIME; + flags |= FS_XFLAG_REALTIME; if (di_flags & XFS_DIFLAG_PREALLOC) - flags |= XFS_XFLAG_PREALLOC; + flags |= FS_XFLAG_PREALLOC; if (di_flags & XFS_DIFLAG_IMMUTABLE) - flags |= XFS_XFLAG_IMMUTABLE; + flags |= FS_XFLAG_IMMUTABLE; if (di_flags & XFS_DIFLAG_APPEND) - flags |= XFS_XFLAG_APPEND; + flags |= FS_XFLAG_APPEND; if (di_flags & XFS_DIFLAG_SYNC) - flags |= XFS_XFLAG_SYNC; + flags |= FS_XFLAG_SYNC; if (di_flags & XFS_DIFLAG_NOATIME) - flags |= XFS_XFLAG_NOATIME; + flags |= FS_XFLAG_NOATIME; if (di_flags & XFS_DIFLAG_NODUMP) - flags |= XFS_XFLAG_NODUMP; + flags |= FS_XFLAG_NODUMP; if (di_flags & XFS_DIFLAG_RTINHERIT) - flags |= XFS_XFLAG_RTINHERIT; + flags |= FS_XFLAG_RTINHERIT; if (di_flags & XFS_DIFLAG_PROJINHERIT) - flags |= XFS_XFLAG_PROJINHERIT; + flags |= FS_XFLAG_PROJINHERIT; if (di_flags & XFS_DIFLAG_NOSYMLINKS) - flags |= XFS_XFLAG_NOSYMLINKS; + flags |= FS_XFLAG_NOSYMLINKS; if (di_flags & XFS_DIFLAG_EXTSIZE) - flags |= XFS_XFLAG_EXTSIZE; + flags |= FS_XFLAG_EXTSIZE; if (di_flags & XFS_DIFLAG_EXTSZINHERIT) - flags |= XFS_XFLAG_EXTSZINHERIT; + flags |= FS_XFLAG_EXTSZINHERIT; if (di_flags & XFS_DIFLAG_NODEFRAG) - flags |= XFS_XFLAG_NODEFRAG; + flags |= FS_XFLAG_NODEFRAG; if (di_flags & XFS_DIFLAG_FILESTREAM) - flags |= XFS_XFLAG_FILESTREAM; + flags |= FS_XFLAG_FILESTREAM; } + if (di_flags2 & XFS_DIFLAG2_ANY) { + if (di_flags2 & XFS_DIFLAG2_DAX) + flags |= FS_XFLAG_DAX; + } + + if (has_attr) + flags |= FS_XFLAG_HASATTR; + return flags; } uint xfs_ip2xflags( - xfs_inode_t *ip) + struct xfs_inode *ip) { - xfs_icdinode_t *dic = &ip->i_d; + struct xfs_icdinode *dic = &ip->i_d; - return _xfs_dic2xflags(dic->di_flags) | - (XFS_IFORK_Q(ip) ? XFS_XFLAG_HASATTR : 0); + return _xfs_dic2xflags(dic->di_flags, dic->di_flags2, XFS_IFORK_Q(ip)); } uint xfs_dic2xflags( - xfs_dinode_t *dip) + struct xfs_dinode *dip) { - return _xfs_dic2xflags(be16_to_cpu(dip->di_flags)) | - (XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0); + return _xfs_dic2xflags(be16_to_cpu(dip->di_flags), + be64_to_cpu(dip->di_flags2), XFS_DFORK_Q(dip)); } /* @@ -862,7 +871,8 @@ xfs_ialloc( case S_IFREG: case S_IFDIR: if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) { - uint di_flags = 0; + uint64_t di_flags2 = 0; + uint di_flags = 0; if (S_ISDIR(mode)) { if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT) @@ -898,7 +908,11 @@ xfs_ialloc( di_flags |= XFS_DIFLAG_NODEFRAG; if (pip->i_d.di_flags & XFS_DIFLAG_FILESTREAM) di_flags |= XFS_DIFLAG_FILESTREAM; + if (pip->i_d.di_flags2 & XFS_DIFLAG2_DAX) + di_flags2 |= XFS_DIFLAG2_DAX; + ip->i_d.di_flags |= di_flags; + ip->i_d.di_flags2 |= di_flags2; } /* FALLTHROUGH */ case S_IFLNK: @@ -1143,7 +1157,6 @@ xfs_create( xfs_bmap_free_t free_list; xfs_fsblock_t first_block; bool unlock_dp_on_error = false; - int committed; prid_t prid; struct xfs_dquot *udqp = NULL; struct xfs_dquot *gdqp = NULL; @@ -1226,7 +1239,7 @@ xfs_create( * pointing to itself. */ error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, - prid, resblks > 0, &ip, &committed); + prid, resblks > 0, &ip, NULL); if (error) goto out_trans_cancel; @@ -1275,7 +1288,7 @@ xfs_create( */ xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp); - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out_bmap_cancel; @@ -1427,7 +1440,6 @@ xfs_link( int error; xfs_bmap_free_t free_list; xfs_fsblock_t first_block; - int committed; int resblks; trace_xfs_link(tdp, target_name); @@ -1502,11 +1514,10 @@ xfs_link( * link transaction goes to disk before returning to * the user. */ - if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) { + if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) xfs_trans_set_sync(tp); - } - error = xfs_bmap_finish (&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) { xfs_bmap_cancel(&free_list); goto error_return; @@ -1555,7 +1566,6 @@ xfs_itruncate_extents( xfs_fileoff_t first_unmap_block; xfs_fileoff_t last_block; xfs_filblks_t unmap_len; - int committed; int error = 0; int done = 0; @@ -1601,9 +1611,7 @@ xfs_itruncate_extents( * Duplicate the transaction that has the permanent * reservation and commit the old transaction. */ - error = xfs_bmap_finish(&tp, &free_list, &committed); - if (committed) - xfs_trans_ijoin(tp, ip, 0); + error = xfs_bmap_finish(&tp, &free_list, ip); if (error) goto out_bmap_cancel; @@ -1774,7 +1782,6 @@ xfs_inactive_ifree( { xfs_bmap_free_t free_list; xfs_fsblock_t first_block; - int committed; struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; int error; @@ -1841,7 +1848,7 @@ xfs_inactive_ifree( * Just ignore errors at this point. There is nothing we can do except * to try to keep going. Make sure it's not a silent error. */ - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) { xfs_notice(mp, "%s: xfs_bmap_finish returned error %d", __func__, error); @@ -2523,7 +2530,6 @@ xfs_remove( int error = 0; xfs_bmap_free_t free_list; xfs_fsblock_t first_block; - int committed; uint resblks; trace_xfs_remove(dp, name); @@ -2624,7 +2630,7 @@ xfs_remove( if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) xfs_trans_set_sync(tp); - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out_bmap_cancel; @@ -2701,7 +2707,6 @@ xfs_finish_rename( struct xfs_trans *tp, struct xfs_bmap_free *free_list) { - int committed = 0; int error; /* @@ -2711,7 +2716,7 @@ xfs_finish_rename( if (tp->t_mountp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) xfs_trans_set_sync(tp); - error = xfs_bmap_finish(&tp, free_list, &committed); + error = xfs_bmap_finish(&tp, free_list, NULL); if (error) { xfs_bmap_cancel(free_list); xfs_trans_cancel(tp); diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index d42738dee..478d04e07 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -859,25 +859,25 @@ xfs_merge_ioc_xflags( unsigned int xflags = start; if (flags & FS_IMMUTABLE_FL) - xflags |= XFS_XFLAG_IMMUTABLE; + xflags |= FS_XFLAG_IMMUTABLE; else - xflags &= ~XFS_XFLAG_IMMUTABLE; + xflags &= ~FS_XFLAG_IMMUTABLE; if (flags & FS_APPEND_FL) - xflags |= XFS_XFLAG_APPEND; + xflags |= FS_XFLAG_APPEND; else - xflags &= ~XFS_XFLAG_APPEND; + xflags &= ~FS_XFLAG_APPEND; if (flags & FS_SYNC_FL) - xflags |= XFS_XFLAG_SYNC; + xflags |= FS_XFLAG_SYNC; else - xflags &= ~XFS_XFLAG_SYNC; + xflags &= ~FS_XFLAG_SYNC; if (flags & FS_NOATIME_FL) - xflags |= XFS_XFLAG_NOATIME; + xflags |= FS_XFLAG_NOATIME; else - xflags &= ~XFS_XFLAG_NOATIME; + xflags &= ~FS_XFLAG_NOATIME; if (flags & FS_NODUMP_FL) - xflags |= XFS_XFLAG_NODUMP; + xflags |= FS_XFLAG_NODUMP; else - xflags &= ~XFS_XFLAG_NODUMP; + xflags &= ~FS_XFLAG_NODUMP; return xflags; } @@ -945,40 +945,51 @@ xfs_set_diflags( unsigned int xflags) { unsigned int di_flags; + uint64_t di_flags2; /* can't set PREALLOC this way, just preserve it */ di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); - if (xflags & XFS_XFLAG_IMMUTABLE) + if (xflags & FS_XFLAG_IMMUTABLE) di_flags |= XFS_DIFLAG_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) + if (xflags & FS_XFLAG_APPEND) di_flags |= XFS_DIFLAG_APPEND; - if (xflags & XFS_XFLAG_SYNC) + if (xflags & FS_XFLAG_SYNC) di_flags |= XFS_DIFLAG_SYNC; - if (xflags & XFS_XFLAG_NOATIME) + if (xflags & FS_XFLAG_NOATIME) di_flags |= XFS_DIFLAG_NOATIME; - if (xflags & XFS_XFLAG_NODUMP) + if (xflags & FS_XFLAG_NODUMP) di_flags |= XFS_DIFLAG_NODUMP; - if (xflags & XFS_XFLAG_NODEFRAG) + if (xflags & FS_XFLAG_NODEFRAG) di_flags |= XFS_DIFLAG_NODEFRAG; - if (xflags & XFS_XFLAG_FILESTREAM) + if (xflags & FS_XFLAG_FILESTREAM) di_flags |= XFS_DIFLAG_FILESTREAM; if (S_ISDIR(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_RTINHERIT) + if (xflags & FS_XFLAG_RTINHERIT) di_flags |= XFS_DIFLAG_RTINHERIT; - if (xflags & XFS_XFLAG_NOSYMLINKS) + if (xflags & FS_XFLAG_NOSYMLINKS) di_flags |= XFS_DIFLAG_NOSYMLINKS; - if (xflags & XFS_XFLAG_EXTSZINHERIT) + if (xflags & FS_XFLAG_EXTSZINHERIT) di_flags |= XFS_DIFLAG_EXTSZINHERIT; - if (xflags & XFS_XFLAG_PROJINHERIT) + if (xflags & FS_XFLAG_PROJINHERIT) di_flags |= XFS_DIFLAG_PROJINHERIT; } else if (S_ISREG(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_REALTIME) + if (xflags & FS_XFLAG_REALTIME) di_flags |= XFS_DIFLAG_REALTIME; - if (xflags & XFS_XFLAG_EXTSIZE) + if (xflags & FS_XFLAG_EXTSIZE) di_flags |= XFS_DIFLAG_EXTSIZE; } - ip->i_d.di_flags = di_flags; + + /* diflags2 only valid for v3 inodes. */ + if (ip->i_d.di_version < 3) + return; + + di_flags2 = 0; + if (xflags & FS_XFLAG_DAX) + di_flags2 |= XFS_DIFLAG2_DAX; + + ip->i_d.di_flags2 = di_flags2; + } STATIC void @@ -988,22 +999,27 @@ xfs_diflags_to_linux( struct inode *inode = VFS_I(ip); unsigned int xflags = xfs_ip2xflags(ip); - if (xflags & XFS_XFLAG_IMMUTABLE) + if (xflags & FS_XFLAG_IMMUTABLE) inode->i_flags |= S_IMMUTABLE; else inode->i_flags &= ~S_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) + if (xflags & FS_XFLAG_APPEND) inode->i_flags |= S_APPEND; else inode->i_flags &= ~S_APPEND; - if (xflags & XFS_XFLAG_SYNC) + if (xflags & FS_XFLAG_SYNC) inode->i_flags |= S_SYNC; else inode->i_flags &= ~S_SYNC; - if (xflags & XFS_XFLAG_NOATIME) + if (xflags & FS_XFLAG_NOATIME) inode->i_flags |= S_NOATIME; else inode->i_flags &= ~S_NOATIME; + if (xflags & FS_XFLAG_DAX) + inode->i_flags |= S_DAX; + else + inode->i_flags &= ~S_DAX; + } static int @@ -1016,11 +1032,11 @@ xfs_ioctl_setattr_xflags( /* Can't change realtime flag if any extents are allocated. */ if ((ip->i_d.di_nextents || ip->i_delayed_blks) && - XFS_IS_REALTIME_INODE(ip) != (fa->fsx_xflags & XFS_XFLAG_REALTIME)) + XFS_IS_REALTIME_INODE(ip) != (fa->fsx_xflags & FS_XFLAG_REALTIME)) return -EINVAL; /* If realtime flag is set then must have realtime device */ - if (fa->fsx_xflags & XFS_XFLAG_REALTIME) { + if (fa->fsx_xflags & FS_XFLAG_REALTIME) { if (mp->m_sb.sb_rblocks == 0 || mp->m_sb.sb_rextsize == 0 || (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) return -EINVAL; @@ -1031,7 +1047,7 @@ xfs_ioctl_setattr_xflags( * we have appropriate permission. */ if (((ip->i_d.di_flags & (XFS_DIFLAG_IMMUTABLE | XFS_DIFLAG_APPEND)) || - (fa->fsx_xflags & (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && + (fa->fsx_xflags & (FS_XFLAG_IMMUTABLE | FS_XFLAG_APPEND))) && !capable(CAP_LINUX_IMMUTABLE)) return -EPERM; @@ -1095,8 +1111,8 @@ out_cancel: * extent size hint validation is somewhat cumbersome. Rules are: * * 1. extent size hint is only valid for directories and regular files - * 2. XFS_XFLAG_EXTSIZE is only valid for regular files - * 3. XFS_XFLAG_EXTSZINHERIT is only valid for directories. + * 2. FS_XFLAG_EXTSIZE is only valid for regular files + * 3. FS_XFLAG_EXTSZINHERIT is only valid for directories. * 4. can only be changed on regular files if no extents are allocated * 5. can be changed on directories at any time * 6. extsize hint of 0 turns off hints, clears inode flags. @@ -1112,10 +1128,10 @@ xfs_ioctl_setattr_check_extsize( { struct xfs_mount *mp = ip->i_mount; - if ((fa->fsx_xflags & XFS_XFLAG_EXTSIZE) && !S_ISREG(ip->i_d.di_mode)) + if ((fa->fsx_xflags & FS_XFLAG_EXTSIZE) && !S_ISREG(ip->i_d.di_mode)) return -EINVAL; - if ((fa->fsx_xflags & XFS_XFLAG_EXTSZINHERIT) && + if ((fa->fsx_xflags & FS_XFLAG_EXTSZINHERIT) && !S_ISDIR(ip->i_d.di_mode)) return -EINVAL; @@ -1132,7 +1148,7 @@ xfs_ioctl_setattr_check_extsize( return -EINVAL; if (XFS_IS_REALTIME_INODE(ip) || - (fa->fsx_xflags & XFS_XFLAG_REALTIME)) { + (fa->fsx_xflags & FS_XFLAG_REALTIME)) { size = mp->m_sb.sb_rextsize << mp->m_sb.sb_blocklog; } else { size = mp->m_sb.sb_blocksize; @@ -1143,7 +1159,7 @@ xfs_ioctl_setattr_check_extsize( if (fa->fsx_extsize % size) return -EINVAL; } else - fa->fsx_xflags &= ~(XFS_XFLAG_EXTSIZE | XFS_XFLAG_EXTSZINHERIT); + fa->fsx_xflags &= ~(FS_XFLAG_EXTSIZE | FS_XFLAG_EXTSZINHERIT); return 0; } @@ -1168,7 +1184,7 @@ xfs_ioctl_setattr_check_projid( if (xfs_get_projid(ip) != fa->fsx_projid) return -EINVAL; - if ((fa->fsx_xflags & XFS_XFLAG_PROJINHERIT) != + if ((fa->fsx_xflags & FS_XFLAG_PROJINHERIT) != (ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)) return -EINVAL; diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index f4f5b43cf..d81bdc080 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -129,7 +129,6 @@ xfs_iomap_write_direct( xfs_trans_t *tp; xfs_bmap_free_t free_list; uint qblocks, resblks, resrtextents; - int committed; int error; int lockmode; int bmapi_flags = XFS_BMAPI_PREALLOC; @@ -203,15 +202,20 @@ xfs_iomap_write_direct( * this outside the transaction context, but if we commit and then crash * we may not have zeroed the blocks and this will be exposed on * recovery of the allocation. Hence we must zero before commit. + * * Further, if we are mapping unwritten extents here, we need to zero * and convert them to written so that we don't need an unwritten extent * callback for DAX. This also means that we need to be able to dip into - * the reserve block pool if there is no space left but we need to do - * unwritten extent conversion. + * the reserve block pool for bmbt block allocation if there is no space + * left but we need to do unwritten extent conversion. */ + if (IS_DAX(VFS_I(ip))) { bmapi_flags = XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO; - tp->t_flags |= XFS_TRANS_RESERVE; + if (ISUNWRITTEN(imap)) { + tp->t_flags |= XFS_TRANS_RESERVE; + resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1; + } } error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, resblks, resrtextents); @@ -247,7 +251,7 @@ xfs_iomap_write_direct( /* * Complete the transaction */ - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out_bmap_cancel; @@ -693,7 +697,7 @@ xfs_iomap_write_allocate( xfs_bmap_free_t free_list; xfs_filblks_t count_fsb; xfs_trans_t *tp; - int nimaps, committed; + int nimaps; int error = 0; int nres; @@ -794,7 +798,7 @@ xfs_iomap_write_allocate( if (error) goto trans_cancel; - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto trans_cancel; @@ -852,7 +856,6 @@ xfs_iomap_write_unwritten( xfs_bmap_free_t free_list; xfs_fsize_t i_size; uint resblks; - int committed; int error; trace_xfs_unwritten_convert(ip, offset, count); @@ -924,7 +927,7 @@ xfs_iomap_write_unwritten( xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); } - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto error_on_bmapi_transaction; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 245268a0c..76b71a1c6 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -414,13 +414,17 @@ xfs_vn_rename( * uio is kmalloced for this reason... */ STATIC const char * -xfs_vn_follow_link( +xfs_vn_get_link( struct dentry *dentry, - void **cookie) + struct inode *inode, + struct delayed_call *done) { char *link; int error = -ENOMEM; + if (!dentry) + return ERR_PTR(-ECHILD); + link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); if (!link) goto out_err; @@ -429,7 +433,8 @@ xfs_vn_follow_link( if (unlikely(error)) goto out_kfree; - return *cookie = link; + set_delayed_call(done, kfree_link, link); + return link; out_kfree: kfree(link); @@ -1172,8 +1177,7 @@ static const struct inode_operations xfs_dir_ci_inode_operations = { static const struct inode_operations xfs_symlink_inode_operations = { .readlink = generic_readlink, - .follow_link = xfs_vn_follow_link, - .put_link = kfree_put_link, + .get_link = xfs_vn_get_link, .getattr = xfs_vn_getattr, .setattr = xfs_vn_setattr, .setxattr = generic_setxattr, @@ -1201,8 +1205,8 @@ xfs_diflags_to_iflags( inode->i_flags |= S_SYNC; if (flags & XFS_DIFLAG_NOATIME) inode->i_flags |= S_NOATIME; - /* XXX: Also needs an on-disk per inode flag! */ - if (ip->i_mount->m_flags & XFS_MOUNT_DAX) + if (ip->i_mount->m_flags & XFS_MOUNT_DAX || + ip->i_d.di_flags2 & XFS_DIFLAG2_DAX) inode->i_flags |= S_DAX; } diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index f52c72a1a..9c9a1c9bc 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1188,10 +1188,16 @@ xlog_iodone(xfs_buf_t *bp) int aborted = 0; /* - * Race to shutdown the filesystem if we see an error. + * Race to shutdown the filesystem if we see an error or the iclog is in + * IOABORT state. The IOABORT state is only set in DEBUG mode to inject + * CRC errors into log recovery. */ - if (XFS_TEST_ERROR(bp->b_error, l->l_mp, - XFS_ERRTAG_IODONE_IOERR, XFS_RANDOM_IODONE_IOERR)) { + if (XFS_TEST_ERROR(bp->b_error, l->l_mp, XFS_ERRTAG_IODONE_IOERR, + XFS_RANDOM_IODONE_IOERR) || + iclog->ic_state & XLOG_STATE_IOABORT) { + if (iclog->ic_state & XLOG_STATE_IOABORT) + iclog->ic_state &= ~XLOG_STATE_IOABORT; + xfs_buf_ioerror_alert(bp, __func__); xfs_buf_stale(bp); xfs_force_shutdown(l->l_mp, SHUTDOWN_LOG_IO_ERROR); @@ -1838,6 +1844,23 @@ xlog_sync( /* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header, iclog->ic_datap, size); +#ifdef DEBUG + /* + * Intentionally corrupt the log record CRC based on the error injection + * frequency, if defined. This facilitates testing log recovery in the + * event of torn writes. Hence, set the IOABORT state to abort the log + * write on I/O completion and shutdown the fs. The subsequent mount + * detects the bad CRC and attempts to recover. + */ + if (log->l_badcrc_factor && + (prandom_u32() % log->l_badcrc_factor == 0)) { + iclog->ic_header.h_crc &= 0xAAAAAAAA; + iclog->ic_state |= XLOG_STATE_IOABORT; + xfs_warn(log->l_mp, + "Intentionally corrupted log record at LSN 0x%llx. Shutdown imminent.", + be64_to_cpu(iclog->ic_header.h_lsn)); + } +#endif bp->b_io_length = BTOBB(count); bp->b_fspriv = iclog; @@ -2045,12 +2068,14 @@ xlog_print_tic_res( "QM_DQCLUSTER", "QM_QINOCREATE", "QM_QUOTAOFF_END", - "SB_UNIT", "FSYNC_TS", "GROWFSRT_ALLOC", "GROWFSRT_ZERO", "GROWFSRT_FREE", - "SWAPEXT" + "SWAPEXT", + "CHECKPOINT", + "ICREATE", + "CREATE_TMPFILE" }; xfs_warn(mp, "xlog_write: reservation summary:"); @@ -2791,11 +2816,19 @@ xlog_state_do_callback( } } while (!ioerrors && loopdidcallbacks); +#ifdef DEBUG /* - * make one last gasp attempt to see if iclogs are being left in - * limbo.. + * Make one last gasp attempt to see if iclogs are being left in limbo. + * If the above loop finds an iclog earlier than the current iclog and + * in one of the syncing states, the current iclog is put into + * DO_CALLBACK and the callbacks are deferred to the completion of the + * earlier iclog. Walk the iclogs in order and make sure that no iclog + * is in DO_CALLBACK unless an earlier iclog is in one of the syncing + * states. + * + * Note that SYNCING|IOABORT is a valid state so we cannot just check + * for ic_state == SYNCING. */ -#ifdef DEBUG if (funcdidcallbacks) { first_iclog = iclog = log->l_iclog; do { @@ -2810,7 +2843,7 @@ xlog_state_do_callback( * IOERROR - give up hope all ye who enter here */ if (iclog->ic_state == XLOG_STATE_WANT_SYNC || - iclog->ic_state == XLOG_STATE_SYNCING || + iclog->ic_state & XLOG_STATE_SYNCING || iclog->ic_state == XLOG_STATE_DONE_SYNC || iclog->ic_state == XLOG_STATE_IOERROR ) break; diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 8daba7491..ed8896310 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -62,6 +62,7 @@ static inline uint xlog_get_client_id(__be32 i) #define XLOG_STATE_CALLBACK 0x0020 /* Callback functions now */ #define XLOG_STATE_DIRTY 0x0040 /* Dirty IC log, not ready for ACTIVE status*/ #define XLOG_STATE_IOERROR 0x0080 /* IO error happened in sync'ing log */ +#define XLOG_STATE_IOABORT 0x0100 /* force abort on I/O completion (debug) */ #define XLOG_STATE_ALL 0x7FFF /* All possible valid flags */ #define XLOG_STATE_NOTUSED 0x8000 /* This IC log not being used */ @@ -410,6 +411,8 @@ struct xlog { /* The following field are used for debugging; need to hold icloglock */ #ifdef DEBUG void *l_iclog_bak[XLOG_MAX_ICLOGS]; + /* log record crc error injection factor */ + uint32_t l_badcrc_factor; #endif }; diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index c5ecaacdd..be5568839 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -61,6 +61,9 @@ xlog_recover_check_summary( #else #define xlog_recover_check_summary(log) #endif +STATIC int +xlog_do_recovery_pass( + struct xlog *, xfs_daddr_t, xfs_daddr_t, int, xfs_daddr_t *); /* * This structure is used during recovery to record the buf log items which @@ -868,136 +871,365 @@ validate_head: } /* - * Find the sync block number or the tail of the log. - * - * This will be the block number of the last record to have its - * associated buffers synced to disk. Every log record header has - * a sync lsn embedded in it. LSNs hold block numbers, so it is easy - * to get a sync block number. The only concern is to figure out which - * log record header to believe. - * - * The following algorithm uses the log record header with the largest - * lsn. The entire log record does not need to be valid. We only care - * that the header is valid. + * Seek backwards in the log for log record headers. * - * We could speed up search by using current head_blk buffer, but it is not - * available. + * Given a starting log block, walk backwards until we find the provided number + * of records or hit the provided tail block. The return value is the number of + * records encountered or a negative error code. The log block and buffer + * pointer of the last record seen are returned in rblk and rhead respectively. */ STATIC int -xlog_find_tail( +xlog_rseek_logrec_hdr( struct xlog *log, - xfs_daddr_t *head_blk, - xfs_daddr_t *tail_blk) + xfs_daddr_t head_blk, + xfs_daddr_t tail_blk, + int count, + struct xfs_buf *bp, + xfs_daddr_t *rblk, + struct xlog_rec_header **rhead, + bool *wrapped) { - xlog_rec_header_t *rhead; - xlog_op_header_t *op_head; + int i; + int error; + int found = 0; char *offset = NULL; - xfs_buf_t *bp; - int error, i, found; - xfs_daddr_t umount_data_blk; - xfs_daddr_t after_umount_blk; - xfs_lsn_t tail_lsn; - int hblks; + xfs_daddr_t end_blk; - found = 0; + *wrapped = false; /* - * Find previous log record + * Walk backwards from the head block until we hit the tail or the first + * block in the log. */ - if ((error = xlog_find_head(log, head_blk))) - return error; - - bp = xlog_get_bp(log, 1); - if (!bp) - return -ENOMEM; - if (*head_blk == 0) { /* special case */ - error = xlog_bread(log, 0, 1, bp, &offset); + end_blk = head_blk > tail_blk ? tail_blk : 0; + for (i = (int) head_blk - 1; i >= end_blk; i--) { + error = xlog_bread(log, i, 1, bp, &offset); if (error) - goto done; + goto out_error; - if (xlog_get_cycle(offset) == 0) { - *tail_blk = 0; - /* leave all other log inited values alone */ - goto done; + if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { + *rblk = i; + *rhead = (struct xlog_rec_header *) offset; + if (++found == count) + break; } } /* - * Search backwards looking for log record header block + * If we haven't hit the tail block or the log record header count, + * start looking again from the end of the physical log. Note that + * callers can pass head == tail if the tail is not yet known. */ - ASSERT(*head_blk < INT_MAX); - for (i = (int)(*head_blk) - 1; i >= 0; i--) { + if (tail_blk >= head_blk && found != count) { + for (i = log->l_logBBsize - 1; i >= (int) tail_blk; i--) { + error = xlog_bread(log, i, 1, bp, &offset); + if (error) + goto out_error; + + if (*(__be32 *)offset == + cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { + *wrapped = true; + *rblk = i; + *rhead = (struct xlog_rec_header *) offset; + if (++found == count) + break; + } + } + } + + return found; + +out_error: + return error; +} + +/* + * Seek forward in the log for log record headers. + * + * Given head and tail blocks, walk forward from the tail block until we find + * the provided number of records or hit the head block. The return value is the + * number of records encountered or a negative error code. The log block and + * buffer pointer of the last record seen are returned in rblk and rhead + * respectively. + */ +STATIC int +xlog_seek_logrec_hdr( + struct xlog *log, + xfs_daddr_t head_blk, + xfs_daddr_t tail_blk, + int count, + struct xfs_buf *bp, + xfs_daddr_t *rblk, + struct xlog_rec_header **rhead, + bool *wrapped) +{ + int i; + int error; + int found = 0; + char *offset = NULL; + xfs_daddr_t end_blk; + + *wrapped = false; + + /* + * Walk forward from the tail block until we hit the head or the last + * block in the log. + */ + end_blk = head_blk > tail_blk ? head_blk : log->l_logBBsize - 1; + for (i = (int) tail_blk; i <= end_blk; i++) { error = xlog_bread(log, i, 1, bp, &offset); if (error) - goto done; + goto out_error; - if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { - found = 1; - break; + if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { + *rblk = i; + *rhead = (struct xlog_rec_header *) offset; + if (++found == count) + break; } } + /* - * If we haven't found the log record header block, start looking - * again from the end of the physical log. XXXmiken: There should be - * a check here to make sure we didn't search more than N blocks in - * the previous code. + * If we haven't hit the head block or the log record header count, + * start looking again from the start of the physical log. */ - if (!found) { - for (i = log->l_logBBsize - 1; i >= (int)(*head_blk); i--) { + if (tail_blk > head_blk && found != count) { + for (i = 0; i < (int) head_blk; i++) { error = xlog_bread(log, i, 1, bp, &offset); if (error) - goto done; + goto out_error; if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) { - found = 2; - break; + *wrapped = true; + *rblk = i; + *rhead = (struct xlog_rec_header *) offset; + if (++found == count) + break; } } } - if (!found) { - xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__); - xlog_put_bp(bp); - ASSERT(0); - return -EIO; + + return found; + +out_error: + return error; +} + +/* + * Check the log tail for torn writes. This is required when torn writes are + * detected at the head and the head had to be walked back to a previous record. + * The tail of the previous record must now be verified to ensure the torn + * writes didn't corrupt the previous tail. + * + * Return an error if CRC verification fails as recovery cannot proceed. + */ +STATIC int +xlog_verify_tail( + struct xlog *log, + xfs_daddr_t head_blk, + xfs_daddr_t tail_blk) +{ + struct xlog_rec_header *thead; + struct xfs_buf *bp; + xfs_daddr_t first_bad; + int count; + int error = 0; + bool wrapped; + xfs_daddr_t tmp_head; + + bp = xlog_get_bp(log, 1); + if (!bp) + return -ENOMEM; + + /* + * Seek XLOG_MAX_ICLOGS + 1 records past the current tail record to get + * a temporary head block that points after the last possible + * concurrently written record of the tail. + */ + count = xlog_seek_logrec_hdr(log, head_blk, tail_blk, + XLOG_MAX_ICLOGS + 1, bp, &tmp_head, &thead, + &wrapped); + if (count < 0) { + error = count; + goto out; } - /* find blk_no of tail of log */ - rhead = (xlog_rec_header_t *)offset; - *tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn)); + /* + * If the call above didn't find XLOG_MAX_ICLOGS + 1 records, we ran + * into the actual log head. tmp_head points to the start of the record + * so update it to the actual head block. + */ + if (count < XLOG_MAX_ICLOGS + 1) + tmp_head = head_blk; /* - * Reset log values according to the state of the log when we - * crashed. In the case where head_blk == 0, we bump curr_cycle - * one because the next write starts a new cycle rather than - * continuing the cycle of the last good log record. At this - * point we have guaranteed that all partial log records have been - * accounted for. Therefore, we know that the last good log record - * written was complete and ended exactly on the end boundary - * of the physical log. + * We now have a tail and temporary head block that covers at least + * XLOG_MAX_ICLOGS records from the tail. We need to verify that these + * records were completely written. Run a CRC verification pass from + * tail to head and return the result. */ - log->l_prev_block = i; - log->l_curr_block = (int)*head_blk; - log->l_curr_cycle = be32_to_cpu(rhead->h_cycle); - if (found == 2) - log->l_curr_cycle++; - atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn)); - atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn)); - xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle, - BBTOB(log->l_curr_block)); - xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle, - BBTOB(log->l_curr_block)); + error = xlog_do_recovery_pass(log, tmp_head, tail_blk, + XLOG_RECOVER_CRCPASS, &first_bad); + +out: + xlog_put_bp(bp); + return error; +} + +/* + * Detect and trim torn writes from the head of the log. + * + * Storage without sector atomicity guarantees can result in torn writes in the + * log in the event of a crash. Our only means to detect this scenario is via + * CRC verification. While we can't always be certain that CRC verification + * failure is due to a torn write vs. an unrelated corruption, we do know that + * only a certain number (XLOG_MAX_ICLOGS) of log records can be written out at + * one time. Therefore, CRC verify up to XLOG_MAX_ICLOGS records at the head of + * the log and treat failures in this range as torn writes as a matter of + * policy. In the event of CRC failure, the head is walked back to the last good + * record in the log and the tail is updated from that record and verified. + */ +STATIC int +xlog_verify_head( + struct xlog *log, + xfs_daddr_t *head_blk, /* in/out: unverified head */ + xfs_daddr_t *tail_blk, /* out: tail block */ + struct xfs_buf *bp, + xfs_daddr_t *rhead_blk, /* start blk of last record */ + struct xlog_rec_header **rhead, /* ptr to last record */ + bool *wrapped) /* last rec. wraps phys. log */ +{ + struct xlog_rec_header *tmp_rhead; + struct xfs_buf *tmp_bp; + xfs_daddr_t first_bad; + xfs_daddr_t tmp_rhead_blk; + int found; + int error; + bool tmp_wrapped; /* - * Look for unmount record. If we find it, then we know there - * was a clean unmount. Since 'i' could be the last block in - * the physical log, we convert to a log block before comparing - * to the head_blk. + * Check the head of the log for torn writes. Search backwards from the + * head until we hit the tail or the maximum number of log record I/Os + * that could have been in flight at one time. Use a temporary buffer so + * we don't trash the rhead/bp pointers from the caller. + */ + tmp_bp = xlog_get_bp(log, 1); + if (!tmp_bp) + return -ENOMEM; + error = xlog_rseek_logrec_hdr(log, *head_blk, *tail_blk, + XLOG_MAX_ICLOGS, tmp_bp, &tmp_rhead_blk, + &tmp_rhead, &tmp_wrapped); + xlog_put_bp(tmp_bp); + if (error < 0) + return error; + + /* + * Now run a CRC verification pass over the records starting at the + * block found above to the current head. If a CRC failure occurs, the + * log block of the first bad record is saved in first_bad. + */ + error = xlog_do_recovery_pass(log, *head_blk, tmp_rhead_blk, + XLOG_RECOVER_CRCPASS, &first_bad); + if (error == -EFSBADCRC) { + /* + * We've hit a potential torn write. Reset the error and warn + * about it. + */ + error = 0; + xfs_warn(log->l_mp, +"Torn write (CRC failure) detected at log block 0x%llx. Truncating head block from 0x%llx.", + first_bad, *head_blk); + + /* + * Get the header block and buffer pointer for the last good + * record before the bad record. + * + * Note that xlog_find_tail() clears the blocks at the new head + * (i.e., the records with invalid CRC) if the cycle number + * matches the the current cycle. + */ + found = xlog_rseek_logrec_hdr(log, first_bad, *tail_blk, 1, bp, + rhead_blk, rhead, wrapped); + if (found < 0) + return found; + if (found == 0) /* XXX: right thing to do here? */ + return -EIO; + + /* + * Reset the head block to the starting block of the first bad + * log record and set the tail block based on the last good + * record. + * + * Bail out if the updated head/tail match as this indicates + * possible corruption outside of the acceptable + * (XLOG_MAX_ICLOGS) range. This is a job for xfs_repair... + */ + *head_blk = first_bad; + *tail_blk = BLOCK_LSN(be64_to_cpu((*rhead)->h_tail_lsn)); + if (*head_blk == *tail_blk) { + ASSERT(0); + return 0; + } + + /* + * Now verify the tail based on the updated head. This is + * required because the torn writes trimmed from the head could + * have been written over the tail of a previous record. Return + * any errors since recovery cannot proceed if the tail is + * corrupt. + * + * XXX: This leaves a gap in truly robust protection from torn + * writes in the log. If the head is behind the tail, the tail + * pushes forward to create some space and then a crash occurs + * causing the writes into the previous record's tail region to + * tear, log recovery isn't able to recover. + * + * How likely is this to occur? If possible, can we do something + * more intelligent here? Is it safe to push the tail forward if + * we can determine that the tail is within the range of the + * torn write (e.g., the kernel can only overwrite the tail if + * it has actually been pushed forward)? Alternatively, could we + * somehow prevent this condition at runtime? + */ + error = xlog_verify_tail(log, *head_blk, *tail_blk); + } + + return error; +} + +/* + * Check whether the head of the log points to an unmount record. In other + * words, determine whether the log is clean. If so, update the in-core state + * appropriately. + */ +static int +xlog_check_unmount_rec( + struct xlog *log, + xfs_daddr_t *head_blk, + xfs_daddr_t *tail_blk, + struct xlog_rec_header *rhead, + xfs_daddr_t rhead_blk, + struct xfs_buf *bp, + bool *clean) +{ + struct xlog_op_header *op_head; + xfs_daddr_t umount_data_blk; + xfs_daddr_t after_umount_blk; + int hblks; + int error; + char *offset; + + *clean = false; + + /* + * Look for unmount record. If we find it, then we know there was a + * clean unmount. Since 'i' could be the last block in the physical + * log, we convert to a log block before comparing to the head_blk. * - * Save the current tail lsn to use to pass to - * xlog_clear_stale_blocks() below. We won't want to clear the - * unmount record if there is one, so we pass the lsn of the - * unmount record rather than the block after it. + * Save the current tail lsn to use to pass to xlog_clear_stale_blocks() + * below. We won't want to clear the unmount record if there is one, so + * we pass the lsn of the unmount record rather than the block after it. */ if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { int h_size = be32_to_cpu(rhead->h_size); @@ -1014,22 +1246,22 @@ xlog_find_tail( } else { hblks = 1; } - after_umount_blk = (i + hblks + (int) - BTOBB(be32_to_cpu(rhead->h_len))) % log->l_logBBsize; - tail_lsn = atomic64_read(&log->l_tail_lsn); + after_umount_blk = rhead_blk + hblks + BTOBB(be32_to_cpu(rhead->h_len)); + after_umount_blk = do_mod(after_umount_blk, log->l_logBBsize); if (*head_blk == after_umount_blk && be32_to_cpu(rhead->h_num_logops) == 1) { - umount_data_blk = (i + hblks) % log->l_logBBsize; + umount_data_blk = rhead_blk + hblks; + umount_data_blk = do_mod(umount_data_blk, log->l_logBBsize); error = xlog_bread(log, umount_data_blk, 1, bp, &offset); if (error) - goto done; + return error; - op_head = (xlog_op_header_t *)offset; + op_head = (struct xlog_op_header *)offset; if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) { /* - * Set tail and last sync so that newly written - * log records will point recovery to after the - * current unmount record. + * Set tail and last sync so that newly written log + * records will point recovery to after the current + * unmount record. */ xlog_assign_atomic_lsn(&log->l_tail_lsn, log->l_curr_cycle, after_umount_blk); @@ -1037,16 +1269,166 @@ xlog_find_tail( log->l_curr_cycle, after_umount_blk); *tail_blk = after_umount_blk; - /* - * Note that the unmount was clean. If the unmount - * was not clean, we need to know this to rebuild the - * superblock counters from the perag headers if we - * have a filesystem using non-persistent counters. - */ - log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN; + *clean = true; } } + return 0; +} + +static void +xlog_set_state( + struct xlog *log, + xfs_daddr_t head_blk, + struct xlog_rec_header *rhead, + xfs_daddr_t rhead_blk, + bool bump_cycle) +{ + /* + * Reset log values according to the state of the log when we + * crashed. In the case where head_blk == 0, we bump curr_cycle + * one because the next write starts a new cycle rather than + * continuing the cycle of the last good log record. At this + * point we have guaranteed that all partial log records have been + * accounted for. Therefore, we know that the last good log record + * written was complete and ended exactly on the end boundary + * of the physical log. + */ + log->l_prev_block = rhead_blk; + log->l_curr_block = (int)head_blk; + log->l_curr_cycle = be32_to_cpu(rhead->h_cycle); + if (bump_cycle) + log->l_curr_cycle++; + atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn)); + atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn)); + xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle, + BBTOB(log->l_curr_block)); + xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle, + BBTOB(log->l_curr_block)); +} + +/* + * Find the sync block number or the tail of the log. + * + * This will be the block number of the last record to have its + * associated buffers synced to disk. Every log record header has + * a sync lsn embedded in it. LSNs hold block numbers, so it is easy + * to get a sync block number. The only concern is to figure out which + * log record header to believe. + * + * The following algorithm uses the log record header with the largest + * lsn. The entire log record does not need to be valid. We only care + * that the header is valid. + * + * We could speed up search by using current head_blk buffer, but it is not + * available. + */ +STATIC int +xlog_find_tail( + struct xlog *log, + xfs_daddr_t *head_blk, + xfs_daddr_t *tail_blk) +{ + xlog_rec_header_t *rhead; + char *offset = NULL; + xfs_buf_t *bp; + int error; + xfs_daddr_t rhead_blk; + xfs_lsn_t tail_lsn; + bool wrapped = false; + bool clean = false; + + /* + * Find previous log record + */ + if ((error = xlog_find_head(log, head_blk))) + return error; + ASSERT(*head_blk < INT_MAX); + + bp = xlog_get_bp(log, 1); + if (!bp) + return -ENOMEM; + if (*head_blk == 0) { /* special case */ + error = xlog_bread(log, 0, 1, bp, &offset); + if (error) + goto done; + + if (xlog_get_cycle(offset) == 0) { + *tail_blk = 0; + /* leave all other log inited values alone */ + goto done; + } + } + + /* + * Search backwards through the log looking for the log record header + * block. This wraps all the way back around to the head so something is + * seriously wrong if we can't find it. + */ + error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, bp, + &rhead_blk, &rhead, &wrapped); + if (error < 0) + return error; + if (!error) { + xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__); + return -EIO; + } + *tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn)); + + /* + * Set the log state based on the current head record. + */ + xlog_set_state(log, *head_blk, rhead, rhead_blk, wrapped); + tail_lsn = atomic64_read(&log->l_tail_lsn); + + /* + * Look for an unmount record at the head of the log. This sets the log + * state to determine whether recovery is necessary. + */ + error = xlog_check_unmount_rec(log, head_blk, tail_blk, rhead, + rhead_blk, bp, &clean); + if (error) + goto done; + + /* + * Verify the log head if the log is not clean (e.g., we have anything + * but an unmount record at the head). This uses CRC verification to + * detect and trim torn writes. If discovered, CRC failures are + * considered torn writes and the log head is trimmed accordingly. + * + * Note that we can only run CRC verification when the log is dirty + * because there's no guarantee that the log data behind an unmount + * record is compatible with the current architecture. + */ + if (!clean) { + xfs_daddr_t orig_head = *head_blk; + + error = xlog_verify_head(log, head_blk, tail_blk, bp, + &rhead_blk, &rhead, &wrapped); + if (error) + goto done; + + /* update in-core state again if the head changed */ + if (*head_blk != orig_head) { + xlog_set_state(log, *head_blk, rhead, rhead_blk, + wrapped); + tail_lsn = atomic64_read(&log->l_tail_lsn); + error = xlog_check_unmount_rec(log, head_blk, tail_blk, + rhead, rhead_blk, bp, + &clean); + if (error) + goto done; + } + } + + /* + * Note that the unmount was clean. If the unmount was not clean, we + * need to know this to rebuild the superblock counters from the perag + * headers if we have a filesystem using non-persistent counters. + */ + if (clean) + log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN; + /* * Make sure that there are no blocks in front of the head * with the same cycle number as the head. This can happen @@ -3204,6 +3586,7 @@ xlog_recover_dquot_ra_pass2( struct xfs_disk_dquot *recddq; struct xfs_dq_logformat *dq_f; uint type; + int len; if (mp->m_qflags == 0) @@ -3224,8 +3607,12 @@ xlog_recover_dquot_ra_pass2( ASSERT(dq_f); ASSERT(dq_f->qlf_len == 1); - xfs_buf_readahead(mp->m_ddev_targp, dq_f->qlf_blkno, - XFS_FSB_TO_BB(mp, dq_f->qlf_len), NULL); + len = XFS_FSB_TO_BB(mp, dq_f->qlf_len); + if (xlog_peek_buffer_cancelled(log, dq_f->qlf_blkno, len, 0)) + return; + + xfs_buf_readahead(mp->m_ddev_targp, dq_f->qlf_blkno, len, + &xfs_dquot_buf_ra_ops); } STATIC void @@ -4118,25 +4505,68 @@ xlog_recover_process_iunlinks( mp->m_dmevmask = mp_dmevmask; } +STATIC int +xlog_unpack_data( + struct xlog_rec_header *rhead, + char *dp, + struct xlog *log) +{ + int i, j, k; + + for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) && + i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) { + *(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i]; + dp += BBSIZE; + } + + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { + xlog_in_core_2_t *xhdr = (xlog_in_core_2_t *)rhead; + for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { + j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); + k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); + *(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k]; + dp += BBSIZE; + } + } + + return 0; +} + /* - * Upack the log buffer data and crc check it. If the check fails, issue a - * warning if and only if the CRC in the header is non-zero. This makes the - * check an advisory warning, and the zero CRC check will prevent failure - * warnings from being emitted when upgrading the kernel from one that does not - * add CRCs by default. - * - * When filesystems are CRC enabled, this CRC mismatch becomes a fatal log - * corruption failure + * CRC check, unpack and process a log record. */ STATIC int -xlog_unpack_data_crc( +xlog_recover_process( + struct xlog *log, + struct hlist_head rhash[], struct xlog_rec_header *rhead, char *dp, - struct xlog *log) + int pass) { + int error; __le32 crc; crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len)); + + /* + * Nothing else to do if this is a CRC verification pass. Just return + * if this a record with a non-zero crc. Unfortunately, mkfs always + * sets h_crc to 0 so we must consider this valid even on v5 supers. + * Otherwise, return EFSBADCRC on failure so the callers up the stack + * know precisely what failed. + */ + if (pass == XLOG_RECOVER_CRCPASS) { + if (rhead->h_crc && crc != rhead->h_crc) + return -EFSBADCRC; + return 0; + } + + /* + * We're in the normal recovery path. Issue a warning if and only if the + * CRC in the header is non-zero. This is an advisory warning and the + * zero CRC check prevents warnings from being emitted when upgrading + * the kernel from one that does not add CRCs by default. + */ if (crc != rhead->h_crc) { if (rhead->h_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) { xfs_alert(log->l_mp, @@ -4147,47 +4577,18 @@ xlog_unpack_data_crc( } /* - * If we've detected a log record corruption, then we can't - * recover past this point. Abort recovery if we are enforcing - * CRC protection by punting an error back up the stack. + * If the filesystem is CRC enabled, this mismatch becomes a + * fatal log corruption failure. */ if (xfs_sb_version_hascrc(&log->l_mp->m_sb)) return -EFSCORRUPTED; } - return 0; -} - -STATIC int -xlog_unpack_data( - struct xlog_rec_header *rhead, - char *dp, - struct xlog *log) -{ - int i, j, k; - int error; - - error = xlog_unpack_data_crc(rhead, dp, log); + error = xlog_unpack_data(rhead, dp, log); if (error) return error; - for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) && - i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) { - *(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i]; - dp += BBSIZE; - } - - if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { - xlog_in_core_2_t *xhdr = (xlog_in_core_2_t *)rhead; - for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { - j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE); - *(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k]; - dp += BBSIZE; - } - } - - return 0; + return xlog_recover_process_data(log, rhash, rhead, dp, pass); } STATIC int @@ -4239,18 +4640,21 @@ xlog_do_recovery_pass( struct xlog *log, xfs_daddr_t head_blk, xfs_daddr_t tail_blk, - int pass) + int pass, + xfs_daddr_t *first_bad) /* out: first bad log rec */ { xlog_rec_header_t *rhead; xfs_daddr_t blk_no; + xfs_daddr_t rhead_blk; char *offset; xfs_buf_t *hbp, *dbp; - int error = 0, h_size; + int error = 0, h_size, h_len; int bblks, split_bblks; int hblks, split_hblks, wrapped_hblks; struct hlist_head rhash[XLOG_RHASH_SIZE]; ASSERT(head_blk != tail_blk); + rhead_blk = 0; /* * Read the header of the tail block and get the iclog buffer size from @@ -4274,7 +4678,31 @@ xlog_do_recovery_pass( error = xlog_valid_rec_header(log, rhead, tail_blk); if (error) goto bread_err1; + + /* + * xfsprogs has a bug where record length is based on lsunit but + * h_size (iclog size) is hardcoded to 32k. Now that we + * unconditionally CRC verify the unmount record, this means the + * log buffer can be too small for the record and cause an + * overrun. + * + * Detect this condition here. Use lsunit for the buffer size as + * long as this looks like the mkfs case. Otherwise, return an + * error to avoid a buffer overrun. + */ h_size = be32_to_cpu(rhead->h_size); + h_len = be32_to_cpu(rhead->h_len); + if (h_len > h_size) { + if (h_len <= log->l_mp->m_logbsize && + be32_to_cpu(rhead->h_num_logops) == 1) { + xfs_warn(log->l_mp, + "invalid iclog size (%d bytes), using lsunit (%d bytes)", + h_size, log->l_mp->m_logbsize); + h_size = log->l_mp->m_logbsize; + } else + return -EFSCORRUPTED; + } + if ((be32_to_cpu(rhead->h_version) & XLOG_VERSION_2) && (h_size > XLOG_HEADER_CYCLE_SIZE)) { hblks = h_size / XLOG_HEADER_CYCLE_SIZE; @@ -4301,7 +4729,7 @@ xlog_do_recovery_pass( } memset(rhash, 0, sizeof(rhash)); - blk_no = tail_blk; + blk_no = rhead_blk = tail_blk; if (tail_blk > head_blk) { /* * Perform recovery around the end of the physical log. @@ -4408,19 +4836,18 @@ xlog_do_recovery_pass( goto bread_err2; } - error = xlog_unpack_data(rhead, offset, log); + error = xlog_recover_process(log, rhash, rhead, offset, + pass); if (error) goto bread_err2; - error = xlog_recover_process_data(log, rhash, - rhead, offset, pass); - if (error) - goto bread_err2; blk_no += bblks; + rhead_blk = blk_no; } ASSERT(blk_no >= log->l_logBBsize); blk_no -= log->l_logBBsize; + rhead_blk = blk_no; } /* read first part of physical log */ @@ -4441,21 +4868,22 @@ xlog_do_recovery_pass( if (error) goto bread_err2; - error = xlog_unpack_data(rhead, offset, log); + error = xlog_recover_process(log, rhash, rhead, offset, pass); if (error) goto bread_err2; - error = xlog_recover_process_data(log, rhash, - rhead, offset, pass); - if (error) - goto bread_err2; blk_no += bblks + hblks; + rhead_blk = blk_no; } bread_err2: xlog_put_bp(dbp); bread_err1: xlog_put_bp(hbp); + + if (error && first_bad) + *first_bad = rhead_blk; + return error; } @@ -4493,7 +4921,7 @@ xlog_do_log_recovery( INIT_LIST_HEAD(&log->l_buf_cancel_table[i]); error = xlog_do_recovery_pass(log, head_blk, tail_blk, - XLOG_RECOVER_PASS1); + XLOG_RECOVER_PASS1, NULL); if (error != 0) { kmem_free(log->l_buf_cancel_table); log->l_buf_cancel_table = NULL; @@ -4504,7 +4932,7 @@ xlog_do_log_recovery( * When it is complete free the table of buf cancel items. */ error = xlog_do_recovery_pass(log, head_blk, tail_blk, - XLOG_RECOVER_PASS2); + XLOG_RECOVER_PASS2, NULL); #ifdef DEBUG if (!error) { int i; diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c index dc6221942..ade236e90 100644 --- a/fs/xfs/xfs_pnfs.c +++ b/fs/xfs/xfs_pnfs.c @@ -42,11 +42,11 @@ xfs_break_layouts( while ((error = break_layout(inode, false) == -EWOULDBLOCK)) { xfs_iunlock(ip, *iolock); if (with_imutex && (*iolock & XFS_IOLOCK_EXCL)) - mutex_unlock(&inode->i_mutex); + inode_unlock(inode); error = break_layout(inode, true); *iolock = XFS_IOLOCK_EXCL; if (with_imutex) - mutex_lock(&inode->i_mutex); + inode_lock(inode); xfs_ilock(ip, *iolock); } diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index ab1bac6a3..be02a68b2 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -766,7 +766,6 @@ xfs_growfs_rt_alloc( { xfs_fileoff_t bno; /* block number in file */ struct xfs_buf *bp; /* temporary buffer for zeroing */ - int committed; /* transaction committed flag */ xfs_daddr_t d; /* disk block address */ int error; /* error return value */ xfs_fsblock_t firstblock;/* first block allocated in xaction */ @@ -811,7 +810,7 @@ xfs_growfs_rt_alloc( /* * Free any blocks freed up in the transaction, then commit. */ - error = xfs_bmap_finish(&tp, &flist, &committed); + error = xfs_bmap_finish(&tp, &flist, NULL); if (error) goto out_bmap_cancel; error = xfs_trans_commit(tp); diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 36bd8825b..59c9b7bd9 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -137,7 +137,7 @@ static const match_table_t tokens = { }; -STATIC unsigned long +STATIC int suffix_kstrtoint(char *s, unsigned int base, int *res) { int last, shift_left_factor = 0, _res; @@ -1714,8 +1714,8 @@ xfs_init_zones(void) xfs_inode_zone = kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode", - KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD, - xfs_fs_inode_init_once); + KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD | + KM_ZONE_ACCOUNT, xfs_fs_inode_init_once); if (!xfs_inode_zone) goto out_destroy_efi_zone; diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 996481eeb..b44284c1a 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -178,7 +178,6 @@ xfs_symlink( struct xfs_bmap_free free_list; xfs_fsblock_t first_block; bool unlock_dp_on_error = false; - int committed; xfs_fileoff_t first_fsb; xfs_filblks_t fs_blocks; int nmaps; @@ -387,7 +386,7 @@ xfs_symlink( xfs_trans_set_sync(tp); } - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, NULL); if (error) goto out_bmap_cancel; @@ -434,7 +433,6 @@ xfs_inactive_symlink_rmt( struct xfs_inode *ip) { xfs_buf_t *bp; - int committed; int done; int error; xfs_fsblock_t first_block; @@ -510,15 +508,9 @@ xfs_inactive_symlink_rmt( /* * Commit the first transaction. This logs the EFI and the inode. */ - error = xfs_bmap_finish(&tp, &free_list, &committed); + error = xfs_bmap_finish(&tp, &free_list, ip); if (error) goto error_bmap_cancel; - /* - * The transaction must have been committed, since there were - * actually extents freed by xfs_bunmapi. See xfs_bmap_finish. - * The new tp has the extent freeing and EFDs. - */ - ASSERT(committed); /* * The first xact was committed, so add the inode to the new one. * Mark it dirty so it will be logged and moved forward in the log as diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c index ee70f5dec..641d625eb 100644 --- a/fs/xfs/xfs_sysfs.c +++ b/fs/xfs/xfs_sysfs.c @@ -255,11 +255,47 @@ write_grant_head_show( } XFS_SYSFS_ATTR_RO(write_grant_head); +#ifdef DEBUG +STATIC ssize_t +log_badcrc_factor_store( + struct kobject *kobject, + const char *buf, + size_t count) +{ + struct xlog *log = to_xlog(kobject); + int ret; + uint32_t val; + + ret = kstrtouint(buf, 0, &val); + if (ret) + return ret; + + log->l_badcrc_factor = val; + + return count; +} + +STATIC ssize_t +log_badcrc_factor_show( + struct kobject *kobject, + char *buf) +{ + struct xlog *log = to_xlog(kobject); + + return snprintf(buf, PAGE_SIZE, "%d\n", log->l_badcrc_factor); +} + +XFS_SYSFS_ATTR_RW(log_badcrc_factor); +#endif /* DEBUG */ + static struct attribute *xfs_log_attrs[] = { ATTR_LIST(log_head_lsn), ATTR_LIST(log_tail_lsn), ATTR_LIST(reserve_grant_head), ATTR_LIST(write_grant_head), +#ifdef DEBUG + ATTR_LIST(log_badcrc_factor), +#endif NULL, }; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 877079eb0..391d797cb 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -1222,6 +1222,32 @@ DEFINE_PAGE_EVENT(xfs_writepage); DEFINE_PAGE_EVENT(xfs_releasepage); DEFINE_PAGE_EVENT(xfs_invalidatepage); +DECLARE_EVENT_CLASS(xfs_readpage_class, + TP_PROTO(struct inode *inode, int nr_pages), + TP_ARGS(inode, nr_pages), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, nr_pages) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->nr_pages = nr_pages; + ), + TP_printk("dev %d:%d ino 0x%llx nr_pages %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->nr_pages) +) + +#define DEFINE_READPAGE_EVENT(name) \ +DEFINE_EVENT(xfs_readpage_class, name, \ + TP_PROTO(struct inode *inode, int nr_pages), \ + TP_ARGS(inode, nr_pages)) +DEFINE_READPAGE_EVENT(xfs_vm_readpage); +DEFINE_READPAGE_EVENT(xfs_vm_readpages); + DECLARE_EVENT_CLASS(xfs_imap_class, TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, int type, struct xfs_bmbt_irec *irec), diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index ce78534a0..995170194 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -572,12 +572,16 @@ xfs_quota_warn( struct xfs_dquot *dqp, int type) { - /* no warnings for project quotas - we just return ENOSPC later */ + enum quota_type qtype; + if (dqp->dq_flags & XFS_DQ_PROJ) - return; - quota_send_warning(make_kqid(&init_user_ns, - (dqp->dq_flags & XFS_DQ_USER) ? - USRQUOTA : GRPQUOTA, + qtype = PRJQUOTA; + else if (dqp->dq_flags & XFS_DQ_USER) + qtype = USRQUOTA; + else + qtype = GRPQUOTA; + + quota_send_warning(make_kqid(&init_user_ns, qtype, be32_to_cpu(dqp->q_core.d_id)), mp->m_super->s_dev, type); } diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 839b35ca2..110f1d7d8 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -39,9 +39,6 @@ xfs_xattr_get(const struct xattr_handler *handler, struct dentry *dentry, struct xfs_inode *ip = XFS_I(d_inode(dentry)); int error, asize = size; - if (strcmp(name, "") == 0) - return -EINVAL; - /* Convert Linux syscall to XFS internal ATTR flags */ if (!size) { xflags |= ATTR_KERNOVAL; @@ -84,9 +81,6 @@ xfs_xattr_set(const struct xattr_handler *handler, struct dentry *dentry, struct xfs_inode *ip = XFS_I(d_inode(dentry)); int error; - if (strcmp(name, "") == 0) - return -EINVAL; - /* Convert Linux syscall to XFS internal ATTR flags */ if (flags & XATTR_CREATE) xflags |= ATTR_CREATE; @@ -135,47 +129,19 @@ const struct xattr_handler *xfs_xattr_handlers[] = { NULL }; -static unsigned int xfs_xattr_prefix_len(int flags) -{ - if (flags & XFS_ATTR_SECURE) - return sizeof("security"); - else if (flags & XFS_ATTR_ROOT) - return sizeof("trusted"); - else - return sizeof("user"); -} - -static const char *xfs_xattr_prefix(int flags) -{ - if (flags & XFS_ATTR_SECURE) - return xfs_xattr_security_handler.prefix; - else if (flags & XFS_ATTR_ROOT) - return xfs_xattr_trusted_handler.prefix; - else - return xfs_xattr_user_handler.prefix; -} - static int -xfs_xattr_put_listent( +__xfs_xattr_put_listent( struct xfs_attr_list_context *context, - int flags, - unsigned char *name, - int namelen, - int valuelen, - unsigned char *value) + char *prefix, + int prefix_len, + unsigned char *name, + int namelen) { - unsigned int prefix_len = xfs_xattr_prefix_len(flags); char *offset; int arraytop; - ASSERT(context->count >= 0); - - /* - * Only show root namespace entries if we are actually allowed to - * see them. - */ - if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN)) - return 0; + if (!context->alist) + goto compute_size; arraytop = context->count + prefix_len + namelen + 1; if (arraytop > context->firstu) { @@ -183,17 +149,19 @@ xfs_xattr_put_listent( return 1; } offset = (char *)context->alist + context->count; - strncpy(offset, xfs_xattr_prefix(flags), prefix_len); + strncpy(offset, prefix, prefix_len); offset += prefix_len; strncpy(offset, (char *)name, namelen); /* real name */ offset += namelen; *offset = '\0'; + +compute_size: context->count += prefix_len + namelen + 1; return 0; } static int -xfs_xattr_put_listent_sizes( +xfs_xattr_put_listent( struct xfs_attr_list_context *context, int flags, unsigned char *name, @@ -201,24 +169,55 @@ xfs_xattr_put_listent_sizes( int valuelen, unsigned char *value) { - context->count += xfs_xattr_prefix_len(flags) + namelen + 1; - return 0; -} + char *prefix; + int prefix_len; -static int -list_one_attr(const char *name, const size_t len, void *data, - size_t size, ssize_t *result) -{ - char *p = data + *result; + ASSERT(context->count >= 0); - *result += len; - if (!size) - return 0; - if (*result > size) - return -ERANGE; + if (flags & XFS_ATTR_ROOT) { +#ifdef CONFIG_XFS_POSIX_ACL + if (namelen == SGI_ACL_FILE_SIZE && + strncmp(name, SGI_ACL_FILE, + SGI_ACL_FILE_SIZE) == 0) { + int ret = __xfs_xattr_put_listent( + context, XATTR_SYSTEM_PREFIX, + XATTR_SYSTEM_PREFIX_LEN, + XATTR_POSIX_ACL_ACCESS, + strlen(XATTR_POSIX_ACL_ACCESS)); + if (ret) + return ret; + } else if (namelen == SGI_ACL_DEFAULT_SIZE && + strncmp(name, SGI_ACL_DEFAULT, + SGI_ACL_DEFAULT_SIZE) == 0) { + int ret = __xfs_xattr_put_listent( + context, XATTR_SYSTEM_PREFIX, + XATTR_SYSTEM_PREFIX_LEN, + XATTR_POSIX_ACL_DEFAULT, + strlen(XATTR_POSIX_ACL_DEFAULT)); + if (ret) + return ret; + } +#endif - strcpy(p, name); - return 0; + /* + * Only show root namespace entries if we are actually allowed to + * see them. + */ + if (!capable(CAP_SYS_ADMIN)) + return 0; + + prefix = XATTR_TRUSTED_PREFIX; + prefix_len = XATTR_TRUSTED_PREFIX_LEN; + } else if (flags & XFS_ATTR_SECURE) { + prefix = XATTR_SECURITY_PREFIX; + prefix_len = XATTR_SECURITY_PREFIX_LEN; + } else { + prefix = XATTR_USER_PREFIX; + prefix_len = XATTR_USER_PREFIX_LEN; + } + + return __xfs_xattr_put_listent(context, prefix, prefix_len, name, + namelen); } ssize_t @@ -227,7 +226,6 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size) struct xfs_attr_list_context context; struct attrlist_cursor_kern cursor = { 0 }; struct inode *inode = d_inode(dentry); - int error; /* * First read the regular on-disk attributes. @@ -236,37 +234,14 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size) context.dp = XFS_I(inode); context.cursor = &cursor; context.resynch = 1; - context.alist = data; + context.alist = size ? data : NULL; context.bufsize = size; context.firstu = context.bufsize; - - if (size) - context.put_listent = xfs_xattr_put_listent; - else - context.put_listent = xfs_xattr_put_listent_sizes; + context.put_listent = xfs_xattr_put_listent; xfs_attr_list_int(&context); if (context.count < 0) return -ERANGE; - /* - * Then add the two synthetic ACL attributes. - */ - if (posix_acl_access_exists(inode)) { - error = list_one_attr(POSIX_ACL_XATTR_ACCESS, - strlen(POSIX_ACL_XATTR_ACCESS) + 1, - data, size, &context.count); - if (error) - return error; - } - - if (posix_acl_default_exists(inode)) { - error = list_one_attr(POSIX_ACL_XATTR_DEFAULT, - strlen(POSIX_ACL_XATTR_DEFAULT) + 1, - data, size, &context.count); - if (error) - return error; - } - return context.count; } -- cgit v1.2.3-54-g00ecf