diff options
33 files changed, 587 insertions, 329 deletions
@@ -319,6 +319,12 @@ static int dax_insert_mapping(struct inode *inode, struct buffer_head *bh, * @vma: The virtual memory area where the fault occurred * @vmf: The description of the fault * @get_block: The filesystem method used to translate file offsets to blocks + * @complete_unwritten: The filesystem method used to convert unwritten blocks + * to written so the data written to them is exposed. This is required for + * required by write faults for filesystems that will return unwritten + * extent mappings from @get_block, but it is optional for reads as + * dax_insert_mapping() will always zero unwritten blocks. If the fs does + * not support unwritten extents, the it should pass NULL. * * When a page fault occurs, filesystems may call this helper in their * fault handler for DAX files. __dax_fault() assumes the caller has done all @@ -437,8 +443,12 @@ int __dax_fault(struct vm_area_struct *vma, struct vm_fault *vmf, * as for normal BH based IO completions. */ error = dax_insert_mapping(inode, &bh, vma, vmf); - if (buffer_unwritten(&bh)) - complete_unwritten(&bh, !error); + if (buffer_unwritten(&bh)) { + if (complete_unwritten) + complete_unwritten(&bh, !error); + else + WARN_ON_ONCE(!(vmf->flags & FAULT_FLAG_WRITE)); + } out: if (error == -ENOMEM) diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index df6828570e87..a096841bd06c 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -33,6 +33,7 @@ xfs-y += $(addprefix libxfs/, \ xfs_attr.o \ xfs_attr_leaf.o \ xfs_attr_remote.o \ + xfs_bit.o \ xfs_bmap.o \ xfs_bmap_btree.o \ xfs_btree.o \ @@ -63,7 +64,6 @@ xfs-$(CONFIG_XFS_RT) += $(addprefix libxfs/, \ xfs-y += xfs_aops.o \ xfs_attr_inactive.o \ xfs_attr_list.o \ - xfs_bit.o \ xfs_bmap_util.o \ xfs_buf.o \ xfs_dir2_readdir.o \ diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c index eba0d1e91a93..f38f9bd81557 100644 --- a/fs/xfs/libxfs/xfs_attr_remote.c +++ b/fs/xfs/libxfs/xfs_attr_remote.c @@ -159,11 +159,10 @@ xfs_attr3_rmt_write_verify( struct xfs_buf *bp) { struct xfs_mount *mp = bp->b_target->bt_mount; - struct xfs_buf_log_item *bip = bp->b_fspriv; + int blksize = mp->m_attr_geo->blksize; char *ptr; int len; xfs_daddr_t bno; - int blksize = mp->m_attr_geo->blksize; /* no verification of non-crc buffers */ if (!xfs_sb_version_hascrc(&mp->m_sb)) @@ -175,16 +174,22 @@ xfs_attr3_rmt_write_verify( ASSERT(len >= blksize); while (len > 0) { + struct xfs_attr3_rmt_hdr *rmt = (struct xfs_attr3_rmt_hdr *)ptr; + if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) { xfs_buf_ioerror(bp, -EFSCORRUPTED); xfs_verifier_error(bp); return; } - if (bip) { - struct xfs_attr3_rmt_hdr *rmt; - rmt = (struct xfs_attr3_rmt_hdr *)ptr; - rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn); + /* + * Ensure we aren't writing bogus LSNs to disk. See + * xfs_attr3_rmt_hdr_set() for the explanation. + */ + if (rmt->rm_lsn != cpu_to_be64(NULLCOMMITLSN)) { + xfs_buf_ioerror(bp, -EFSCORRUPTED); + xfs_verifier_error(bp); + return; } xfs_update_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF); @@ -221,6 +226,18 @@ xfs_attr3_rmt_hdr_set( rmt->rm_owner = cpu_to_be64(ino); rmt->rm_blkno = cpu_to_be64(bno); + /* + * Remote attribute blocks are written synchronously, so we don't + * have an LSN that we can stamp in them that makes any sense to log + * recovery. To ensure that log recovery handles overwrites of these + * blocks sanely (i.e. once they've been freed and reallocated as some + * other type of metadata) we need to ensure that the LSN has a value + * that tells log recovery to ignore the LSN and overwrite the buffer + * with whatever is in it's log. To do this, we use the magic + * NULLCOMMITLSN to indicate that the LSN is invalid. + */ + rmt->rm_lsn = cpu_to_be64(NULLCOMMITLSN); + return sizeof(struct xfs_attr3_rmt_hdr); } @@ -434,14 +451,21 @@ xfs_attr_rmtval_set( /* * Allocate a single extent, up to the size of the value. + * + * Note that we have to consider this a data allocation as we + * write the remote attribute without logging the contents. + * Hence we must ensure that we aren't using blocks that are on + * the busy list so that we don't overwrite blocks which have + * recently been freed but their transactions are not yet + * committed to disk. If we overwrite the contents of a busy + * extent and then crash then the block may not contain the + * correct metadata after log recovery occurs. */ xfs_bmap_init(args->flist, args->firstblock); nmap = 1; error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno, - blkcnt, - XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA, - args->firstblock, args->total, &map, &nmap, - args->flist); + blkcnt, XFS_BMAPI_ATTRFORK, args->firstblock, + args->total, &map, &nmap, args->flist); if (!error) { error = xfs_bmap_finish(&args->trans, args->flist, &committed); @@ -594,9 +618,8 @@ xfs_attr_rmtval_remove( xfs_bmap_init(args->flist, args->firstblock); error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt, - XFS_BMAPI_ATTRFORK | XFS_BMAPI_METADATA, - 1, args->firstblock, args->flist, - &done); + XFS_BMAPI_ATTRFORK, 1, args->firstblock, + args->flist, &done); if (!error) { error = xfs_bmap_finish(&args->trans, args->flist, &committed); diff --git a/fs/xfs/xfs_bit.c b/fs/xfs/libxfs/xfs_bit.c index 0e8885a59646..0e8885a59646 100644 --- a/fs/xfs/xfs_bit.c +++ b/fs/xfs/libxfs/xfs_bit.c diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 63e05b663380..8e2010d53b07 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5945,6 +5945,7 @@ xfs_bmap_split_extent( return xfs_trans_commit(tp); out: + xfs_bmap_cancel(&free_list); xfs_trans_cancel(tp); return error; } diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index f6a8631dc771..cd2201f5ab52 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -2356,8 +2356,8 @@ xfs_da_shrink_inode( * the last block to the place we want to kill. */ error = xfs_bunmapi(tp, dp, dead_blkno, count, - xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA, - 0, args->firstblock, args->flist, &done); + xfs_bmapi_aflag(w), 0, args->firstblock, + args->flist, &done); if (error == -ENOSPC) { if (w != XFS_DATA_FORK) break; diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c index 42799d88ecc0..9de401d297e5 100644 --- a/fs/xfs/libxfs/xfs_dir2.c +++ b/fs/xfs/libxfs/xfs_dir2.c @@ -677,25 +677,22 @@ xfs_dir2_shrink_inode( mp = dp->i_mount; tp = args->trans; da = xfs_dir2_db_to_da(args->geo, db); - /* - * Unmap the fsblock(s). - */ - if ((error = xfs_bunmapi(tp, dp, da, args->geo->fsbcount, - XFS_BMAPI_METADATA, 0, args->firstblock, args->flist, - &done))) { + + /* Unmap the fsblock(s). */ + error = xfs_bunmapi(tp, dp, da, args->geo->fsbcount, 0, 0, + args->firstblock, args->flist, &done); + if (error) { /* - * ENOSPC actually can happen if we're in a removename with - * no space reservation, and the resulting block removal - * would cause a bmap btree split or conversion from extents - * to btree. This can only happen for un-fragmented - * directory blocks, since you need to be punching out - * the middle of an extent. - * In this case we need to leave the block in the file, - * and not binval it. - * So the block has to be in a consistent empty state - * and appropriately logged. - * We don't free up the buffer, the caller can tell it - * hasn't happened since it got an error back. + * ENOSPC actually can happen if we're in a removename with no + * space reservation, and the resulting block removal would + * cause a bmap btree split or conversion from extents to btree. + * This can only happen for un-fragmented directory blocks, + * since you need to be punching out the middle of an extent. + * In this case we need to leave the block in the file, and not + * binval it. So the block has to be in a consistent empty + * state and appropriately logged. We don't free up the buffer, + * the caller can tell it hasn't happened since it got an error + * back. */ return error; } diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c index 527b7337a43b..6687489412a0 100644 --- a/fs/xfs/libxfs/xfs_dir2_node.c +++ b/fs/xfs/libxfs/xfs_dir2_node.c @@ -1845,8 +1845,7 @@ xfs_dir2_node_addname_int( if (dp->d_ops->db_to_fdb(args->geo, dbno) != fbno) { xfs_alert(mp, - "%s: dir ino %llu needed freesp block %lld for\n" - " data block %lld, got %lld ifbno %llu lastfbno %d", +"%s: dir ino %llu needed freesp block %lld for data block %lld, got %lld ifbno %llu lastfbno %d", __func__, (unsigned long long)dp->i_ino, (long long)dp->d_ops->db_to_fdb( args->geo, dbno), diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c index ce63e0431f3e..54deb2d12ac6 100644 --- a/fs/xfs/libxfs/xfs_ialloc.c +++ b/fs/xfs/libxfs/xfs_ialloc.c @@ -2233,7 +2233,7 @@ xfs_imap_lookup( } xfs_trans_brelse(tp, agbp); - xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); if (error) return error; diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 9c87f66ab929..47425140f343 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -131,10 +131,11 @@ xfs_mount_validate_sb( if (xfs_sb_has_compat_feature(sbp, XFS_SB_FEAT_COMPAT_UNKNOWN)) { xfs_warn(mp, -"Superblock has unknown compatible features (0x%x) enabled.\n" -"Using a more recent kernel is recommended.", +"Superblock has unknown compatible features (0x%x) enabled.", (sbp->sb_features_compat & XFS_SB_FEAT_COMPAT_UNKNOWN)); + xfs_warn(mp, +"Using a more recent kernel is recommended."); } if (xfs_sb_has_ro_compat_feature(sbp, @@ -145,18 +146,21 @@ xfs_mount_validate_sb( XFS_SB_FEAT_RO_COMPAT_UNKNOWN)); if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { xfs_warn(mp, -"Attempted to mount read-only compatible filesystem read-write.\n" +"Attempted to mount read-only compatible filesystem read-write."); + xfs_warn(mp, "Filesystem can only be safely mounted read only."); + return -EINVAL; } } if (xfs_sb_has_incompat_feature(sbp, XFS_SB_FEAT_INCOMPAT_UNKNOWN)) { xfs_warn(mp, -"Superblock has unknown incompatible features (0x%x) enabled.\n" -"Filesystem can not be safely mounted by this kernel.", +"Superblock has unknown incompatible features (0x%x) enabled.", (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_UNKNOWN)); + xfs_warn(mp, +"Filesystem can not be safely mounted by this kernel."); return -EINVAL; } } diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 0f34886cf726..3bf4ad0d19e4 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -67,16 +67,15 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb) */ int /* error */ xfs_bmap_finish( - xfs_trans_t **tp, /* transaction pointer addr */ - xfs_bmap_free_t *flist, /* i/o: list extents to free */ - int *committed) /* xact committed or not */ + struct xfs_trans **tp, /* transaction pointer addr */ + struct xfs_bmap_free *flist, /* i/o: list extents to free */ + int *committed)/* xact committed or not */ { - xfs_efd_log_item_t *efd; /* extent free data */ - xfs_efi_log_item_t *efi; /* extent free intention */ - int error; /* error return value */ - xfs_bmap_free_item_t *free; /* free extent item */ - xfs_mount_t *mp; /* filesystem mount structure */ - xfs_bmap_free_item_t *next; /* next item on free list */ + struct xfs_efd_log_item *efd; /* extent free data */ + struct xfs_efi_log_item *efi; /* extent free intention */ + int error; /* error return value */ + 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) { @@ -88,40 +87,48 @@ xfs_bmap_finish( xfs_trans_log_efi_extent(*tp, efi, free->xbfi_startblock, free->xbfi_blockcount); - error = xfs_trans_roll(tp, NULL); - *committed = 1; - /* - * We have a new transaction, so we should return committed=1, - * even though we're returning an error. - */ - if (error) + error = __xfs_trans_roll(tp, NULL, committed); + if (error) { + /* + * If the transaction was committed, drop the EFD reference + * since we're bailing out of here. The other reference is + * dropped when the EFI hits the AIL. + * + * If the transaction was not committed, the EFI is freed by the + * EFI item unlock handler on abort. Also, we have a new + * transaction so we should return committed=1 even though we're + * returning an error. + */ + 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; + } + /* + * Get an EFD and free each extent in the list, logging to the EFD in + * the process. The remaining bmap free list is cleaned up by the caller + * on error. + */ efd = xfs_trans_get_efd(*tp, efi, flist->xbf_count); for (free = flist->xbf_first; free != NULL; free = next) { next = free->xbfi_next; - if ((error = xfs_free_extent(*tp, free->xbfi_startblock, - free->xbfi_blockcount))) { - /* - * The bmap free list will be cleaned up at a - * higher level. The EFI will be canceled when - * this transaction is aborted. - * Need to force shutdown here to make sure it - * happens, since this transaction may not be - * dirty yet. - */ - mp = (*tp)->t_mountp; - if (!XFS_FORCED_SHUTDOWN(mp)) - xfs_force_shutdown(mp, - (error == -EFSCORRUPTED) ? - SHUTDOWN_CORRUPT_INCORE : - SHUTDOWN_META_IO_ERROR); + + error = xfs_trans_free_extent(*tp, efd, free->xbfi_startblock, + free->xbfi_blockcount); + if (error) return error; - } - xfs_trans_log_efd_extent(*tp, efd, free->xbfi_startblock, - free->xbfi_blockcount); + xfs_bmap_del_free(flist, NULL, free); } + return 0; } @@ -1467,7 +1474,7 @@ xfs_shift_file_space( XFS_DIOSTRAT_SPACE_RES(mp, 0), 0, XFS_QMOPT_RES_REGBLKS); if (error) - goto out; + goto out_trans_cancel; xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); @@ -1481,18 +1488,20 @@ xfs_shift_file_space( &done, stop_fsb, &first_block, &free_list, direction, XFS_BMAP_MAX_SHIFT_EXTENTS); if (error) - goto out; + goto out_bmap_cancel; error = xfs_bmap_finish(&tp, &free_list, &committed); if (error) - goto out; + goto out_bmap_cancel; error = xfs_trans_commit(tp); } return error; -out: +out_bmap_cancel: + xfs_bmap_cancel(&free_list); +out_trans_cancel: xfs_trans_cancel(tp); return error; } diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index a4b7d92e946c..d7dbd8120aaa 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -1533,9 +1533,10 @@ xfs_wait_buftarg( list_del_init(&bp->b_lru); if (bp->b_flags & XBF_WRITE_FAIL) { xfs_alert(btp->bt_mount, -"Corruption Alert: Buffer at block 0x%llx had permanent write failures!\n" -"Please run xfs_repair to determine the extent of the problem.", +"Corruption Alert: Buffer at block 0x%llx had permanent write failures!", (long long)bp->b_bn); + xfs_alert(btp->bt_mount, +"Please run xfs_repair to determine the extent of the problem."); } xfs_buf_rele(bp); } diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 092d652bc03d..919057e0a45b 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -647,11 +647,7 @@ xfs_buf_item_unlock( xfs_buf_item_relse(bp); else if (aborted) { ASSERT(XFS_FORCED_SHUTDOWN(lip->li_mountp)); - if (lip->li_flags & XFS_LI_IN_AIL) { - spin_lock(&lip->li_ailp->xa_lock); - xfs_trans_ail_delete(lip->li_ailp, lip, - SHUTDOWN_LOG_IO_ERROR); - } + xfs_trans_ail_remove(lip, SHUTDOWN_LOG_IO_ERROR); xfs_buf_item_relse(bp); } } diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index b1b26b6a0735..30cb3afb67f0 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -954,12 +954,8 @@ xfs_qm_dqflush( struct xfs_log_item *lip = &dqp->q_logitem.qli_item; dqp->dq_flags &= ~XFS_DQ_DIRTY; - spin_lock(&mp->m_ail->xa_lock); - if (lip->li_flags & XFS_LI_IN_AIL) - xfs_trans_ail_delete(mp->m_ail, lip, - SHUTDOWN_CORRUPT_INCORE); - else - spin_unlock(&mp->m_ail->xa_lock); + xfs_trans_ail_remove(lip, SHUTDOWN_CORRUPT_INCORE); + error = -EIO; goto out_unlock; } diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index adc8f8fdd145..4aa0153214f9 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -47,28 +47,6 @@ xfs_efi_item_free( } /* - * Freeing the efi requires that we remove it from the AIL if it has already - * been placed there. However, the EFI may not yet have been placed in the AIL - * when called by xfs_efi_release() from EFD processing due to the ordering of - * committed vs unpin operations in bulk insert operations. Hence the reference - * count to ensure only the last caller frees the EFI. - */ -STATIC void -__xfs_efi_release( - struct xfs_efi_log_item *efip) -{ - struct xfs_ail *ailp = efip->efi_item.li_ailp; - - if (atomic_dec_and_test(&efip->efi_refcount)) { - spin_lock(&ailp->xa_lock); - /* xfs_trans_ail_delete() drops the AIL lock. */ - xfs_trans_ail_delete(ailp, &efip->efi_item, - SHUTDOWN_LOG_IO_ERROR); - xfs_efi_item_free(efip); - } -} - -/* * This returns the number of iovecs needed to log the given efi item. * We only need 1 iovec for an efi item. It just logs the efi_log_format * structure. @@ -128,12 +106,12 @@ xfs_efi_item_pin( } /* - * While EFIs cannot really be pinned, the unpin operation is the last place at - * which the EFI is manipulated during a transaction. If we are being asked to - * remove the EFI it's because the transaction has been cancelled and by - * definition that means the EFI cannot be in the AIL so remove it from the - * transaction and free it. Otherwise coordinate with xfs_efi_release() - * to determine who gets to free the EFI. + * The unpin operation is the last place an EFI is manipulated in the log. It is + * either inserted in the AIL or aborted in the event of a log I/O error. In + * either case, the EFI transaction has been successfully committed to make it + * this far. Therefore, we expect whoever committed the EFI to either construct + * and commit the EFD or drop the EFD's reference in the event of error. Simply + * drop the log's EFI reference now that the log is done with it. */ STATIC void xfs_efi_item_unpin( @@ -141,15 +119,7 @@ xfs_efi_item_unpin( int remove) { struct xfs_efi_log_item *efip = EFI_ITEM(lip); - - if (remove) { - ASSERT(!(lip->li_flags & XFS_LI_IN_AIL)); - if (lip->li_desc) - xfs_trans_del_item(lip); - xfs_efi_item_free(efip); - return; - } - __xfs_efi_release(efip); + xfs_efi_release(efip); } /* @@ -167,6 +137,11 @@ xfs_efi_item_push( return XFS_ITEM_PINNED; } +/* + * The EFI has been either committed or aborted if the transaction has been + * cancelled. If the transaction was cancelled, an EFD isn't going to be + * constructed and thus we free the EFI here directly. + */ STATIC void xfs_efi_item_unlock( struct xfs_log_item *lip) @@ -301,23 +276,19 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt) } /* - * This is called by the efd item code below to release references to the given - * efi item. Each efd calls this with the number of extents that it has - * logged, and when the sum of these reaches the total number of extents logged - * by this efi item we can free the efi item. + * Freeing the efi requires that we remove it from the AIL if it has already + * been placed there. However, the EFI may not yet have been placed in the AIL + * when called by xfs_efi_release() from EFD processing due to the ordering of + * committed vs unpin operations in bulk insert operations. Hence the reference + * count to ensure only the last caller frees the EFI. */ void -xfs_efi_release(xfs_efi_log_item_t *efip, - uint nextents) +xfs_efi_release( + struct xfs_efi_log_item *efip) { - ASSERT(atomic_read(&efip->efi_next_extent) >= nextents); - if (atomic_sub_and_test(nextents, &efip->efi_next_extent)) { - /* recovery needs us to drop the EFI reference, too */ - if (test_bit(XFS_EFI_RECOVERED, &efip->efi_flags)) - __xfs_efi_release(efip); - - __xfs_efi_release(efip); - /* efip may now have been freed, do not reference it again. */ + if (atomic_dec_and_test(&efip->efi_refcount)) { + xfs_trans_ail_remove(&efip->efi_item, SHUTDOWN_LOG_IO_ERROR); + xfs_efi_item_free(efip); } } @@ -415,20 +386,27 @@ xfs_efd_item_push( return XFS_ITEM_PINNED; } +/* + * The EFD is either committed or aborted if the transaction is cancelled. If + * the transaction is cancelled, drop our reference to the EFI and free the EFD. + */ STATIC void xfs_efd_item_unlock( struct xfs_log_item *lip) { - if (lip->li_flags & XFS_LI_ABORTED) - xfs_efd_item_free(EFD_ITEM(lip)); + struct xfs_efd_log_item *efdp = EFD_ITEM(lip); + + if (lip->li_flags & XFS_LI_ABORTED) { + xfs_efi_release(efdp->efd_efip); + xfs_efd_item_free(efdp); + } } /* - * When the efd item is committed to disk, all we need to do - * is delete our reference to our partner efi item and then - * free ourselves. Since we're freeing ourselves we must - * return -1 to keep the transaction code from further referencing - * this item. + * When the efd item is committed to disk, all we need to do is delete our + * reference to our partner efi item and then free ourselves. Since we're + * freeing ourselves we must return -1 to keep the transaction code from further + * referencing this item. */ STATIC xfs_lsn_t xfs_efd_item_committed( @@ -438,13 +416,14 @@ xfs_efd_item_committed( struct xfs_efd_log_item *efdp = EFD_ITEM(lip); /* - * If we got a log I/O error, it's always the case that the LR with the - * EFI got unpinned and freed before the EFD got aborted. + * Drop the EFI reference regardless of whether the EFD has been + * aborted. Once the EFD transaction is constructed, it is the sole + * responsibility of the EFD to release the EFI (even if the EFI is + * aborted due to log I/O error). */ - if (!(lip->li_flags & XFS_LI_ABORTED)) - xfs_efi_release(efdp->efd_efip, efdp->efd_format.efd_nextents); - + xfs_efi_release(efdp->efd_efip); xfs_efd_item_free(efdp); + return (xfs_lsn_t)-1; } diff --git a/fs/xfs/xfs_extfree_item.h b/fs/xfs/xfs_extfree_item.h index 0ffbce32d569..8fa8651705e1 100644 --- a/fs/xfs/xfs_extfree_item.h +++ b/fs/xfs/xfs_extfree_item.h @@ -39,9 +39,28 @@ struct kmem_zone; * "extent free done" log item described below. * * The EFI is reference counted so that it is not freed prior to both the EFI - * and EFD being committed and unpinned. This ensures that when the last - * reference goes away the EFI will always be in the AIL as it has been - * unpinned, regardless of whether the EFD is processed before or after the EFI. + * and EFD being committed and unpinned. This ensures the EFI is inserted into + * the AIL even in the event of out of order EFI/EFD processing. In other words, + * an EFI is born with two references: + * + * 1.) an EFI held reference to track EFI AIL insertion + * 2.) an EFD held reference to track EFD commit + * + * On allocation, both references are the responsibility of the caller. Once the + * EFI is added to and dirtied in a transaction, ownership of reference one + * transfers to the transaction. The reference is dropped once the EFI is + * inserted to the AIL or in the event of failure along the way (e.g., commit + * failure, log I/O error, etc.). Note that the caller remains responsible for + * the EFD reference under all circumstances to this point. The caller has no + * means to detect failure once the transaction is committed, however. + * Therefore, an EFD is required after this point, even in the event of + * unrelated failure. + * + * Once an EFD is allocated and dirtied in a transaction, reference two + * transfers to the transaction. The EFD reference is dropped once it reaches + * the unpin handler. Similar to the EFI, the reference also drops in the event + * of commit failure or log I/O errors. Note that the EFD is not inserted in the + * AIL, so at this point both the EFI and EFD are freed. */ typedef struct xfs_efi_log_item { xfs_log_item_t efi_item; @@ -77,5 +96,6 @@ xfs_efd_log_item_t *xfs_efd_init(struct xfs_mount *, xfs_efi_log_item_t *, int xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt); void xfs_efi_item_free(xfs_efi_log_item_t *); +void xfs_efi_release(struct xfs_efi_log_item *); #endif /* __XFS_EXTFREE_ITEM_H__ */ diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 2d91ab066370..de2c2376242b 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1521,18 +1521,27 @@ xfs_filemap_fault( struct vm_area_struct *vma, struct vm_fault *vmf) { - struct xfs_inode *ip = XFS_I(file_inode(vma->vm_file)); + struct inode *inode = file_inode(vma->vm_file); int ret; - trace_xfs_filemap_fault(ip); + trace_xfs_filemap_fault(XFS_I(inode)); /* DAX can shortcut the normal fault path on write faults! */ - if ((vmf->flags & FAULT_FLAG_WRITE) && IS_DAX(VFS_I(ip))) + if ((vmf->flags & FAULT_FLAG_WRITE) && IS_DAX(inode)) return xfs_filemap_page_mkwrite(vma, vmf); - xfs_ilock(ip, XFS_MMAPLOCK_SHARED); - ret = filemap_fault(vma, vmf); - xfs_iunlock(ip, XFS_MMAPLOCK_SHARED); + xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED); + if (IS_DAX(inode)) { + /* + * we do not want to trigger unwritten extent conversion on read + * faults - that is unnecessary overhead and would also require + * changes to xfs_get_blocks_direct() to map unwritten extent + * ioend for conversion on read-only mappings. + */ + ret = __dax_fault(vma, vmf, xfs_get_blocks_direct, NULL); + } else + ret = filemap_fault(vma, vmf); + xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED); return ret; } diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index aa00ccc0bd78..30555f8fd44b 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1221,11 +1221,8 @@ xfs_create( */ error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, prid, resblks > 0, &ip, &committed); - if (error) { - if (error == -ENOSPC) - goto out_trans_cancel; + if (error) goto out_trans_cancel; - } /* * Now we join the directory inode to the transaction. We do not do it @@ -1364,11 +1361,8 @@ xfs_create_tmpfile( error = xfs_dir_ialloc(&tp, dp, mode, 1, 0, prid, resblks > 0, &ip, NULL); - if (error) { - if (error == -ENOSPC) - goto out_trans_cancel; + if (error) goto out_trans_cancel; - } if (mp->m_flags & XFS_MOUNT_WSYNC) xfs_trans_set_sync(tp); @@ -1838,14 +1832,15 @@ xfs_inactive_ifree( xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_ICOUNT, -1); /* - * 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. + * 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); - if (error) + if (error) { xfs_notice(mp, "%s: xfs_bmap_finish returned error %d", __func__, error); + xfs_bmap_cancel(&free_list); + } error = xfs_trans_commit(tp); if (error) xfs_notice(mp, "%s: xfs_trans_commit returned error %d", diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index bf13a5a7e2f4..62bd80f4edd9 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -703,17 +703,10 @@ xfs_iflush_abort( xfs_inode_log_item_t *iip = ip->i_itemp; if (iip) { - struct xfs_ail *ailp = iip->ili_item.li_ailp; if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { - spin_lock(&ailp->xa_lock); - if (iip->ili_item.li_flags & XFS_LI_IN_AIL) { - /* xfs_trans_ail_delete() drops the AIL lock. */ - xfs_trans_ail_delete(ailp, &iip->ili_item, - stale ? - SHUTDOWN_LOG_IO_ERROR : + xfs_trans_ail_remove(&iip->ili_item, + stale ? SHUTDOWN_LOG_IO_ERROR : SHUTDOWN_CORRUPT_INCORE); - } else - spin_unlock(&ailp->xa_lock); } iip->ili_logged = 0; /* diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index f41b0c3fddab..930ebd86beba 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -473,7 +473,8 @@ xfs_bulkstat( * pending error, then we are done. */ del_cursor: - xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR); + xfs_btree_del_cursor(cur, error ? + XFS_BTREE_ERROR : XFS_BTREE_NOERROR); xfs_buf_relse(agbp); if (error) break; diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 08d4fe46f0fa..aaadee0969c9 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -668,9 +668,9 @@ xfs_log_mount( ASSERT(0); goto out_free_log; } + xfs_crit(mp, "Log size out of supported range."); xfs_crit(mp, -"Log size out of supported range. Continuing onwards, but if log hangs are\n" -"experienced then please report this message in the bug report."); +"Continuing onwards, but if log hangs are experienced then please report this message in the bug report."); } /* @@ -700,6 +700,7 @@ xfs_log_mount( if (error) { xfs_warn(mp, "log mount/recovery failed: error %d", error); + xlog_recover_cancel(mp->m_log); goto out_destroy_ail; } } @@ -740,18 +741,35 @@ out: * it. */ int -xfs_log_mount_finish(xfs_mount_t *mp) +xfs_log_mount_finish( + struct xfs_mount *mp) { int error = 0; - if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) { - error = xlog_recover_finish(mp->m_log); - if (!error) - xfs_log_work_queue(mp); - } else { + if (mp->m_flags & XFS_MOUNT_NORECOVERY) { ASSERT(mp->m_flags & XFS_MOUNT_RDONLY); + return 0; } + error = xlog_recover_finish(mp->m_log); + if (!error) + xfs_log_work_queue(mp); + + return error; +} + +/* + * The mount has failed. Cancel the recovery if it hasn't completed and destroy + * the log. + */ +int +xfs_log_mount_cancel( + struct xfs_mount *mp) +{ + int error; + + error = xlog_recover_cancel(mp->m_log); + xfs_log_unmount(mp); return error; } @@ -1142,11 +1160,13 @@ xlog_space_left( * In this case we just want to return the size of the * log as the amount of space left. */ + xfs_alert(log->l_mp, "xlog_space_left: head behind tail"); xfs_alert(log->l_mp, - "xlog_space_left: head behind tail\n" - " tail_cycle = %d, tail_bytes = %d\n" - " GH cycle = %d, GH bytes = %d", - tail_cycle, tail_bytes, head_cycle, head_bytes); + " tail_cycle = %d, tail_bytes = %d", + tail_cycle, tail_bytes); + xfs_alert(log->l_mp, + " GH cycle = %d, GH bytes = %d", + head_cycle, head_bytes); ASSERT(0); free_bytes = log->l_logsize; } @@ -1652,8 +1672,13 @@ xlog_cksum( if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { union xlog_in_core2 *xhdr = (union xlog_in_core2 *)rhead; int i; + int xheads; + + xheads = size / XLOG_HEADER_CYCLE_SIZE; + if (size % XLOG_HEADER_CYCLE_SIZE) + xheads++; - for (i = 1; i < log->l_iclog_heads; i++) { + for (i = 1; i < xheads; i++) { crc = crc32c(crc, &xhdr[i].hic_xheader, sizeof(struct xlog_rec_ext_header)); } @@ -2028,26 +2053,24 @@ xlog_print_tic_res( "SWAPEXT" }; - xfs_warn(mp, - "xlog_write: reservation summary:\n" - " trans type = %s (%u)\n" - " unit res = %d bytes\n" - " current res = %d bytes\n" - " total reg = %u bytes (o/flow = %u bytes)\n" - " ophdrs = %u (ophdr space = %u bytes)\n" - " ophdr + reg = %u bytes\n" - " num regions = %u", - ((ticket->t_trans_type <= 0 || - ticket->t_trans_type > XFS_TRANS_TYPE_MAX) ? + xfs_warn(mp, "xlog_write: reservation summary:"); + xfs_warn(mp, " trans type = %s (%u)", + ((ticket->t_trans_type <= 0 || + ticket->t_trans_type > XFS_TRANS_TYPE_MAX) ? "bad-trans-type" : trans_type_str[ticket->t_trans_type-1]), - ticket->t_trans_type, - ticket->t_unit_res, - ticket->t_curr_res, - ticket->t_res_arr_sum, ticket->t_res_o_flow, - ticket->t_res_num_ophdrs, ophdr_spc, - ticket->t_res_arr_sum + - ticket->t_res_o_flow + ophdr_spc, - ticket->t_res_num); + ticket->t_trans_type); + xfs_warn(mp, " unit res = %d bytes", + ticket->t_unit_res); + xfs_warn(mp, " current res = %d bytes", + ticket->t_curr_res); + xfs_warn(mp, " total reg = %u bytes (o/flow = %u bytes)", + ticket->t_res_arr_sum, ticket->t_res_o_flow); + xfs_warn(mp, " ophdrs = %u (ophdr space = %u bytes)", + ticket->t_res_num_ophdrs, ophdr_spc); + xfs_warn(mp, " ophdr + reg = %u bytes", + ticket->t_res_arr_sum + ticket->t_res_o_flow + ophdr_spc); + xfs_warn(mp, " num regions = %u", + ticket->t_res_num); for (i = 0; i < ticket->t_res_num; i++) { uint r_type = ticket->t_res_arr[i].r_type; diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index fa27aaec72cb..09d91d3166cd 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -147,6 +147,7 @@ int xfs_log_mount(struct xfs_mount *mp, xfs_daddr_t start_block, int num_bblocks); int xfs_log_mount_finish(struct xfs_mount *mp); +int xfs_log_mount_cancel(struct xfs_mount *); xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp); xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp); void xfs_log_space_wake(struct xfs_mount *mp); diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index abc2ccbff739..4e7649351f5a 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -307,7 +307,13 @@ xlog_cil_insert_items( if (!(lidp->lid_flags & XFS_LID_DIRTY)) continue; - list_move_tail(&lip->li_cil, &cil->xc_cil); + /* + * Only move the item if it isn't already at the tail. This is + * to prevent a transient list_empty() state when reinserting + * an item that is already the only item in the CIL. + */ + if (!list_is_last(&lip->li_cil, &cil->xc_cil)) + list_move_tail(&lip->li_cil, &cil->xc_cil); } /* account for space used by new iovec headers */ diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index 1c87c8abfbed..950f3f94720c 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -426,6 +426,8 @@ xlog_recover( extern int xlog_recover_finish( struct xlog *log); +extern int +xlog_recover_cancel(struct xlog *); extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, char *dp, int size); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 86c3de477a9d..512a0945d52a 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1886,9 +1886,14 @@ xlog_recover_get_buf_lsn( uuid = &((struct xfs_dir3_blk_hdr *)blk)->uuid; break; case XFS_ATTR3_RMT_MAGIC: - lsn = be64_to_cpu(((struct xfs_attr3_rmt_hdr *)blk)->rm_lsn); - uuid = &((struct xfs_attr3_rmt_hdr *)blk)->rm_uuid; - break; + /* + * Remote attr blocks are written synchronously, rather than + * being logged. That means they do not contain a valid LSN + * (i.e. transactionally ordered) in them, and hence any time we + * see a buffer to replay over the top of a remote attribute + * block we should simply do so. + */ + goto recover_immediately; case XFS_SB_MAGIC: /* * superblock uuids are magic. We may or may not have a @@ -2938,16 +2943,16 @@ xlog_recover_efi_pass2( struct xlog_recover_item *item, xfs_lsn_t lsn) { - int error; - xfs_mount_t *mp = log->l_mp; - xfs_efi_log_item_t *efip; - xfs_efi_log_format_t *efi_formatp; + int error; + struct xfs_mount *mp = log->l_mp; + struct xfs_efi_log_item *efip; + struct xfs_efi_log_format *efi_formatp; efi_formatp = item->ri_buf[0].i_addr; efip = xfs_efi_init(mp, efi_formatp->efi_nextents); - if ((error = xfs_efi_copy_format(&(item->ri_buf[0]), - &(efip->efi_format)))) { + error = xfs_efi_copy_format(&item->ri_buf[0], &efip->efi_format); + if (error) { xfs_efi_item_free(efip); return error; } @@ -2955,20 +2960,23 @@ xlog_recover_efi_pass2( spin_lock(&log->l_ailp->xa_lock); /* - * xfs_trans_ail_update() drops the AIL lock. + * The EFI has two references. One for the EFD and one for EFI to ensure + * it makes it into the AIL. Insert the EFI into the AIL directly and + * drop the EFI reference. Note that xfs_trans_ail_update() drops the + * AIL lock. */ xfs_trans_ail_update(log->l_ailp, &efip->efi_item, lsn); + xfs_efi_release(efip); return 0; } /* - * This routine is called when an efd format structure is found in - * a committed transaction in the log. It's purpose is to cancel - * the corresponding efi if it was still in the log. To do this - * it searches the AIL for the efi with an id equal to that in the - * efd format structure. If we find it, we remove the efi from the - * AIL and free it. + * This routine is called when an EFD format structure is found in a committed + * transaction in the log. Its purpose is to cancel the corresponding EFI if it + * was still in the log. To do this it searches the AIL for the EFI with an id + * equal to that in the EFD format structure. If we find it we drop the EFD + * reference, which removes the EFI from the AIL and frees it. */ STATIC int xlog_recover_efd_pass2( @@ -2990,8 +2998,8 @@ xlog_recover_efd_pass2( efi_id = efd_formatp->efd_efi_id; /* - * Search for the efi with the id in the efd format structure - * in the AIL. + * Search for the EFI with the id in the EFD format structure in the + * AIL. */ spin_lock(&ailp->xa_lock); lip = xfs_trans_ail_cursor_first(ailp, &cur, 0); @@ -3000,18 +3008,18 @@ xlog_recover_efd_pass2( efip = (xfs_efi_log_item_t *)lip; if (efip->efi_format.efi_id == efi_id) { /* - * xfs_trans_ail_delete() drops the - * AIL lock. + * Drop the EFD reference to the EFI. This + * removes the EFI from the AIL and frees it. */ - xfs_trans_ail_delete(ailp, lip, - SHUTDOWN_CORRUPT_INCORE); - xfs_efi_item_free(efip); + spin_unlock(&ailp->xa_lock); + xfs_efi_release(efip); spin_lock(&ailp->xa_lock); break; } } lip = xfs_trans_ail_cursor_next(ailp, &cur); } + xfs_trans_ail_cursor_done(&cur); spin_unlock(&ailp->xa_lock); @@ -3039,6 +3047,11 @@ xlog_recover_do_icreate_pass2( unsigned int count; unsigned int isize; xfs_agblock_t length; + int blks_per_cluster; + int bb_per_cluster; + int cancel_count; + int nbufs; + int i; icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr; if (icl->icl_type != XFS_LI_ICREATE) { @@ -3097,22 +3110,45 @@ xlog_recover_do_icreate_pass2( } /* - * Inode buffers can be freed. Do not replay the inode initialisation as - * we could be overwriting something written after this inode buffer was - * cancelled. + * The icreate transaction can cover multiple cluster buffers and these + * buffers could have been freed and reused. Check the individual + * buffers for cancellation so we don't overwrite anything written after + * a cancellation. + */ + blks_per_cluster = xfs_icluster_size_fsb(mp); + bb_per_cluster = XFS_FSB_TO_BB(mp, blks_per_cluster); + nbufs = length / blks_per_cluster; + for (i = 0, cancel_count = 0; i < nbufs; i++) { + xfs_daddr_t daddr; + + daddr = XFS_AGB_TO_DADDR(mp, agno, + agbno + i * blks_per_cluster); + if (xlog_check_buffer_cancelled(log, daddr, bb_per_cluster, 0)) + cancel_count++; + } + + /* + * We currently only use icreate for a single allocation at a time. This + * means we should expect either all or none of the buffers to be + * cancelled. Be conservative and skip replay if at least one buffer is + * cancelled, but warn the user that something is awry if the buffers + * are not consistent. * - * XXX: we need to iterate all buffers and only init those that are not - * cancelled. I think that a more fine grained factoring of - * xfs_ialloc_inode_init may be appropriate here to enable this to be - * done easily. + * XXX: This must be refined to only skip cancelled clusters once we use + * icreate for multiple chunk allocations. */ - if (xlog_check_buffer_cancelled(log, - XFS_AGB_TO_DADDR(mp, agno, agbno), length, 0)) + ASSERT(!cancel_count || cancel_count == nbufs); + if (cancel_count) { + if (cancel_count != nbufs) + xfs_warn(mp, + "WARNING: partial inode chunk cancellation, skipped icreate."); + trace_xfs_log_recover_icreate_cancel(log, icl); return 0; + } - xfs_ialloc_inode_init(mp, NULL, buffer_list, count, agno, agbno, length, - be32_to_cpu(icl->icl_gen)); - return 0; + trace_xfs_log_recover_icreate_recover(log, icl); + return xfs_ialloc_inode_init(mp, NULL, buffer_list, count, agno, agbno, + length, be32_to_cpu(icl->icl_gen)); } STATIC void @@ -3390,14 +3426,24 @@ xlog_recover_add_to_cont_trans( char *ptr, *old_ptr; int old_len; + /* + * If the transaction is empty, the header was split across this and the + * previous record. Copy the rest of the header. + */ if (list_empty(&trans->r_itemq)) { - /* finish copying rest of trans header */ + ASSERT(len < sizeof(struct xfs_trans_header)); + if (len > sizeof(struct xfs_trans_header)) { + xfs_warn(log->l_mp, "%s: bad header length", __func__); + return -EIO; + } + xlog_recover_add_item(&trans->r_itemq); ptr = (char *)&trans->r_theader + - sizeof(xfs_trans_header_t) - len; + sizeof(struct xfs_trans_header) - len; memcpy(ptr, dp, len); return 0; } + /* take the tail entry */ item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list); @@ -3446,7 +3492,19 @@ xlog_recover_add_to_trans( ASSERT(0); return -EIO; } - if (len == sizeof(xfs_trans_header_t)) + + if (len > sizeof(struct xfs_trans_header)) { + xfs_warn(log->l_mp, "%s: bad header length", __func__); + ASSERT(0); + return -EIO; + } + + /* + * The transaction header can be arbitrarily split across op + * records. If we don't have the whole thing here, copy what we + * do have and handle the rest in the next record. + */ + if (len == sizeof(struct xfs_trans_header)) xlog_recover_add_item(&trans->r_itemq); memcpy(&trans->r_theader, dp, len); return 0; @@ -3749,7 +3807,7 @@ xlog_recover_process_efi( * free the memory associated with it. */ set_bit(XFS_EFI_RECOVERED, &efip->efi_flags); - xfs_efi_release(efip, efip->efi_format.efi_nextents); + xfs_efi_release(efip); return -EIO; } } @@ -3762,11 +3820,11 @@ xlog_recover_process_efi( for (i = 0; i < efip->efi_format.efi_nextents; i++) { extp = &(efip->efi_format.efi_extents[i]); - error = xfs_free_extent(tp, extp->ext_start, extp->ext_len); + error = xfs_trans_free_extent(tp, efdp, extp->ext_start, + extp->ext_len); if (error) goto abort_error; - xfs_trans_log_efd_extent(tp, efdp, extp->ext_start, - extp->ext_len); + } set_bit(XFS_EFI_RECOVERED, &efip->efi_flags); @@ -3798,10 +3856,10 @@ abort_error: */ STATIC int xlog_recover_process_efis( - struct xlog *log) + struct xlog *log) { - xfs_log_item_t *lip; - xfs_efi_log_item_t *efip; + struct xfs_log_item *lip; + struct xfs_efi_log_item *efip; int error = 0; struct xfs_ail_cursor cur; struct xfs_ail *ailp; @@ -3825,7 +3883,7 @@ xlog_recover_process_efis( /* * Skip EFIs that we've already processed. */ - efip = (xfs_efi_log_item_t *)lip; + efip = container_of(lip, struct xfs_efi_log_item, efi_item); if (test_bit(XFS_EFI_RECOVERED, &efip->efi_flags)) { lip = xfs_trans_ail_cursor_next(ailp, &cur); continue; @@ -3845,6 +3903,50 @@ out: } /* + * A cancel occurs when the mount has failed and we're bailing out. Release all + * pending EFIs so they don't pin the AIL. + */ +STATIC int +xlog_recover_cancel_efis( + struct xlog *log) +{ + struct xfs_log_item *lip; + struct xfs_efi_log_item *efip; + int error = 0; + struct xfs_ail_cursor cur; + struct xfs_ail *ailp; + + ailp = log->l_ailp; + spin_lock(&ailp->xa_lock); + lip = xfs_trans_ail_cursor_first(ailp, &cur, 0); + while (lip != NULL) { + /* + * We're done when we see something other than an EFI. + * There should be no EFIs left in the AIL now. + */ + if (lip->li_type != XFS_LI_EFI) { +#ifdef DEBUG + for (; lip; lip = xfs_trans_ail_cursor_next(ailp, &cur)) + ASSERT(lip->li_type != XFS_LI_EFI); +#endif + break; + } + + efip = container_of(lip, struct xfs_efi_log_item, efi_item); + + spin_unlock(&ailp->xa_lock); + xfs_efi_release(efip); + spin_lock(&ailp->xa_lock); + + lip = xfs_trans_ail_cursor_next(ailp, &cur); + } + + xfs_trans_ail_cursor_done(&cur); + spin_unlock(&ailp->xa_lock); + return error; +} + +/* * This routine performs a transaction to null out a bad inode pointer * in an agi unlinked inode hash bucket. */ @@ -4537,11 +4639,13 @@ xlog_recover( xfs_sb_has_incompat_log_feature(&log->l_mp->m_sb, XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN)) { xfs_warn(log->l_mp, -"Superblock has unknown incompatible log features (0x%x) enabled.\n" -"The log can not be fully and/or safely recovered by this kernel.\n" -"Please recover the log on a kernel that supports the unknown features.", +"Superblock has unknown incompatible log features (0x%x) enabled.", (log->l_mp->m_sb.sb_features_log_incompat & XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN)); + xfs_warn(log->l_mp, +"The log can not be fully and/or safely recovered by this kernel."); + xfs_warn(log->l_mp, +"Please recover the log on a kernel that supports the unknown features."); return -EINVAL; } @@ -4617,6 +4721,17 @@ xlog_recover_finish( return 0; } +int +xlog_recover_cancel( + struct xlog *log) +{ + int error = 0; + + if (log->l_flags & XLOG_RECOVERY_NEEDED) + error = xlog_recover_cancel_efis(log); + + return error; +} #if defined(DEBUG) /* diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 461e791efad7..bf92e0c037c7 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -615,14 +615,14 @@ xfs_default_resblks(xfs_mount_t *mp) */ int xfs_mountfs( - xfs_mount_t *mp) + struct xfs_mount *mp) { - xfs_sb_t *sbp = &(mp->m_sb); - xfs_inode_t *rip; - __uint64_t resblks; - uint quotamount = 0; - uint quotaflags = 0; - int error = 0; + struct xfs_sb *sbp = &(mp->m_sb); + struct xfs_inode *rip; + __uint64_t resblks; + uint quotamount = 0; + uint quotaflags = 0; + int error = 0; xfs_sb_mount_common(mp, sbp); @@ -799,7 +799,9 @@ xfs_mountfs( } /* - * log's mount-time initialization. Perform 1st part recovery if needed + * Log's mount-time initialization. The first part of recovery can place + * some items on the AIL, to be handled when recovery is finished or + * cancelled. */ error = xfs_log_mount(mp, mp->m_logdev_targp, XFS_FSB_TO_DADDR(mp, sbp->sb_logstart), @@ -910,9 +912,9 @@ xfs_mountfs( } /* - * Finish recovering the file system. This part needed to be - * delayed until after the root and real-time bitmap inodes - * were consistently read in. + * Finish recovering the file system. This part needed to be delayed + * until after the root and real-time bitmap inodes were consistently + * read in. */ error = xfs_log_mount_finish(mp); if (error) { @@ -955,8 +957,10 @@ xfs_mountfs( xfs_rtunmount_inodes(mp); out_rele_rip: IRELE(rip); + cancel_delayed_work_sync(&mp->m_reclaim_work); + xfs_reclaim_inodes(mp, SYNC_WAIT); out_log_dealloc: - xfs_log_unmount(mp); + xfs_log_mount_cancel(mp); out_fail_wait: if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) xfs_wait_buftarg(mp->m_logdev_targp); diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index f4e8c06eee26..ab1bac6a3a1c 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -757,31 +757,30 @@ xfs_rtallocate_extent_size( /* * Allocate space to the bitmap or summary file, and zero it, for growfs. */ -STATIC int /* error */ +STATIC int xfs_growfs_rt_alloc( - xfs_mount_t *mp, /* file system mount point */ - xfs_extlen_t oblocks, /* old count of blocks */ - xfs_extlen_t nblocks, /* new count of blocks */ - xfs_inode_t *ip) /* inode (bitmap/summary) */ + struct xfs_mount *mp, /* file system mount point */ + xfs_extlen_t oblocks, /* old count of blocks */ + xfs_extlen_t nblocks, /* new count of blocks */ + struct xfs_inode *ip) /* inode (bitmap/summary) */ { - xfs_fileoff_t bno; /* block number in file */ - xfs_buf_t *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 */ - xfs_bmap_free_t flist; /* list of freed blocks */ - xfs_fsblock_t fsbno; /* filesystem block for bno */ - xfs_bmbt_irec_t map; /* block map output */ - int nmap; /* number of block maps */ - int resblks; /* space reservation */ + 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 */ + struct xfs_bmap_free flist; /* list of freed blocks */ + xfs_fsblock_t fsbno; /* filesystem block for bno */ + struct xfs_bmbt_irec map; /* block map output */ + int nmap; /* number of block maps */ + int resblks; /* space reservation */ + struct xfs_trans *tp; /* * Allocate space to the file, as necessary. */ while (oblocks < nblocks) { - xfs_trans_t *tp; - tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFSRT_ALLOC); resblks = XFS_GROWFSRT_SPACE_RES(mp, nblocks - oblocks); /* @@ -790,7 +789,7 @@ xfs_growfs_rt_alloc( error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtalloc, resblks, 0); if (error) - goto error_cancel; + goto out_trans_cancel; /* * Lock the inode. */ @@ -808,16 +807,16 @@ xfs_growfs_rt_alloc( if (!error && nmap < 1) error = -ENOSPC; if (error) - goto error_cancel; + goto out_bmap_cancel; /* * Free any blocks freed up in the transaction, then commit. */ error = xfs_bmap_finish(&tp, &flist, &committed); if (error) - goto error_cancel; + goto out_bmap_cancel; error = xfs_trans_commit(tp); if (error) - goto error; + return error; /* * Now we need to clear the allocated blocks. * Do this one block per transaction, to keep it simple. @@ -832,7 +831,7 @@ xfs_growfs_rt_alloc( error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtzero, 0, 0); if (error) - goto error_cancel; + goto out_trans_cancel; /* * Lock the bitmap inode. */ @@ -846,9 +845,7 @@ xfs_growfs_rt_alloc( mp->m_bsize, 0); if (bp == NULL) { error = -EIO; -error_cancel: - xfs_trans_cancel(tp); - goto error; + goto out_trans_cancel; } memset(bp->b_addr, 0, mp->m_sb.sb_blocksize); xfs_trans_log_buf(tp, bp, 0, mp->m_sb.sb_blocksize - 1); @@ -857,16 +854,20 @@ error_cancel: */ error = xfs_trans_commit(tp); if (error) - goto error; + return error; } /* * Go on to the next extent, if any. */ oblocks = map.br_startoff + map.br_blockcount; } + return 0; -error: +out_bmap_cancel: + xfs_bmap_cancel(&flist); +out_trans_cancel: + xfs_trans_cancel(tp); return error; } diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 7a01486eff06..996481eeb491 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -502,7 +502,7 @@ xfs_inactive_symlink_rmt( /* * Unmap the dead block(s) to the free_list. */ - error = xfs_bunmapi(tp, ip, 0, size, XFS_BMAPI_METADATA, nmaps, + error = xfs_bunmapi(tp, ip, 0, size, 0, nmaps, &first_block, &free_list, &done); if (error) goto error_bmap_cancel; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 8d916d33d93d..9aeeb21bc3d0 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -2089,6 +2089,40 @@ DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover); DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel); DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip); +DECLARE_EVENT_CLASS(xfs_log_recover_icreate_item_class, + TP_PROTO(struct xlog *log, struct xfs_icreate_log *in_f), + TP_ARGS(log, in_f), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(unsigned int, count) + __field(unsigned int, isize) + __field(xfs_agblock_t, length) + __field(unsigned int, gen) + ), + TP_fast_assign( + __entry->dev = log->l_mp->m_super->s_dev; + __entry->agno = be32_to_cpu(in_f->icl_ag); + __entry->agbno = be32_to_cpu(in_f->icl_agbno); + __entry->count = be32_to_cpu(in_f->icl_count); + __entry->isize = be32_to_cpu(in_f->icl_isize); + __entry->length = be32_to_cpu(in_f->icl_length); + __entry->gen = be32_to_cpu(in_f->icl_gen); + ), + TP_printk("dev %d:%d agno %u agbno %u count %u isize %u length %u " + "gen %u", MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, __entry->agbno, __entry->count, __entry->isize, + __entry->length, __entry->gen) +) +#define DEFINE_LOG_RECOVER_ICREATE_ITEM(name) \ +DEFINE_EVENT(xfs_log_recover_icreate_item_class, name, \ + TP_PROTO(struct xlog *log, struct xfs_icreate_log *in_f), \ + TP_ARGS(log, in_f)) + +DEFINE_LOG_RECOVER_ICREATE_ITEM(xfs_log_recover_icreate_cancel); +DEFINE_LOG_RECOVER_ICREATE_ITEM(xfs_log_recover_icreate_recover); + DECLARE_EVENT_CLASS(xfs_discard_class, TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, xfs_extlen_t len), diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 0582a27107d4..a0ab1dae9c31 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1019,9 +1019,10 @@ xfs_trans_cancel( * chunk we've been working on and get a new transaction to continue. */ int -xfs_trans_roll( +__xfs_trans_roll( struct xfs_trans **tpp, - struct xfs_inode *dp) + struct xfs_inode *dp, + int *committed) { struct xfs_trans *trans; struct xfs_trans_res tres; @@ -1052,6 +1053,7 @@ xfs_trans_roll( if (error) return error; + *committed = 1; trans = *tpp; /* @@ -1074,3 +1076,12 @@ xfs_trans_roll( xfs_trans_ijoin(trans, dp, 0); return 0; } + +int +xfs_trans_roll( + struct xfs_trans **tpp, + struct xfs_inode *dp) +{ + int committed = 0; + return __xfs_trans_roll(tpp, dp, &committed); +} diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 3b21b4e5e467..4643070d7cae 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -213,7 +213,6 @@ void xfs_trans_ijoin(struct xfs_trans *, struct xfs_inode *, uint); void xfs_trans_log_buf(xfs_trans_t *, struct xfs_buf *, uint, uint); void xfs_trans_log_inode(xfs_trans_t *, struct xfs_inode *, uint); struct xfs_efi_log_item *xfs_trans_get_efi(xfs_trans_t *, uint); -void xfs_efi_release(struct xfs_efi_log_item *, uint); void xfs_trans_log_efi_extent(xfs_trans_t *, struct xfs_efi_log_item *, xfs_fsblock_t, @@ -221,11 +220,11 @@ void xfs_trans_log_efi_extent(xfs_trans_t *, struct xfs_efd_log_item *xfs_trans_get_efd(xfs_trans_t *, struct xfs_efi_log_item *, uint); -void xfs_trans_log_efd_extent(xfs_trans_t *, - struct xfs_efd_log_item *, - xfs_fsblock_t, - xfs_extlen_t); +int xfs_trans_free_extent(struct xfs_trans *, + struct xfs_efd_log_item *, xfs_fsblock_t, + xfs_extlen_t); int xfs_trans_commit(struct xfs_trans *); +int __xfs_trans_roll(struct xfs_trans **, struct xfs_inode *, int *); int xfs_trans_roll(struct xfs_trans **, struct xfs_inode *); void xfs_trans_cancel(xfs_trans_t *); int xfs_trans_ail_init(struct xfs_mount *); diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c index 284397dd7990..a96ae540eb62 100644 --- a/fs/xfs/xfs_trans_extfree.c +++ b/fs/xfs/xfs_trans_extfree.c @@ -25,6 +25,7 @@ #include "xfs_trans.h" #include "xfs_trans_priv.h" #include "xfs_extfree_item.h" +#include "xfs_alloc.h" /* * This routine is called to allocate an "extent free intention" @@ -108,19 +109,30 @@ xfs_trans_get_efd(xfs_trans_t *tp, } /* - * This routine is called to indicate that the described - * extent is to be logged as having been freed. It should - * be called once for each extent freed. + * Free an extent and log it to the EFD. Note that the transaction is marked + * dirty regardless of whether the extent free succeeds or fails to support the + * EFI/EFD lifecycle rules. */ -void -xfs_trans_log_efd_extent(xfs_trans_t *tp, - xfs_efd_log_item_t *efdp, - xfs_fsblock_t start_block, - xfs_extlen_t ext_len) +int +xfs_trans_free_extent( + struct xfs_trans *tp, + struct xfs_efd_log_item *efdp, + xfs_fsblock_t start_block, + xfs_extlen_t ext_len) { uint next_extent; - xfs_extent_t *extp; + struct xfs_extent *extp; + int error; + error = xfs_free_extent(tp, start_block, ext_len); + + /* + * Mark the transaction dirty, even on error. This ensures the + * transaction is aborted, which: + * + * 1.) releases the EFI and frees the EFD + * 2.) shuts down the filesystem + */ tp->t_flags |= XFS_TRANS_DIRTY; efdp->efd_item.li_desc->lid_flags |= XFS_LID_DIRTY; @@ -130,4 +142,6 @@ xfs_trans_log_efd_extent(xfs_trans_t *tp, extp->ext_start = start_block; extp->ext_len = ext_len; efdp->efd_next_extent++; + + return error; } diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 1b736294558a..49931b72da8a 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -119,6 +119,21 @@ xfs_trans_ail_delete( xfs_trans_ail_delete_bulk(ailp, &lip, 1, shutdown_type); } +static inline void +xfs_trans_ail_remove( + struct xfs_log_item *lip, + int shutdown_type) +{ + struct xfs_ail *ailp = lip->li_ailp; + + spin_lock(&ailp->xa_lock); + /* xfs_trans_ail_delete() drops the AIL lock */ + if (lip->li_flags & XFS_LI_IN_AIL) + xfs_trans_ail_delete(ailp, lip, shutdown_type); + else + spin_unlock(&ailp->xa_lock); +} + void xfs_ail_push(struct xfs_ail *, xfs_lsn_t); void xfs_ail_push_all(struct xfs_ail *); void xfs_ail_push_all_sync(struct xfs_ail *); |