diff options
Diffstat (limited to 'fs')
38 files changed, 647 insertions, 372 deletions
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 5a5d325a3935..634608d2a6d0 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -147,14 +147,12 @@ struct btrfs_inode { * the btrfs file release call will add this inode to the * ordered operations list so that we make sure to flush out any * new data the application may have written before commit. - * - * yes, its silly to have a single bitflag, but we might grow more - * of these. */ unsigned ordered_data_close:1; unsigned orphan_meta_reserved:1; unsigned dummy_inode:1; unsigned in_defrag:1; + unsigned delalloc_meta_reserved:1; /* * always compress this one file diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 3a1b939c9ae2..5b163572e0ca 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -617,12 +617,14 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_root *root, static int btrfs_delayed_inode_reserve_metadata( struct btrfs_trans_handle *trans, struct btrfs_root *root, + struct inode *inode, struct btrfs_delayed_node *node) { struct btrfs_block_rsv *src_rsv; struct btrfs_block_rsv *dst_rsv; u64 num_bytes; int ret; + int release = false; src_rsv = trans->block_rsv; dst_rsv = &root->fs_info->delayed_block_rsv; @@ -652,12 +654,65 @@ static int btrfs_delayed_inode_reserve_metadata( if (!ret) node->bytes_reserved = num_bytes; return ret; + } else if (src_rsv == &root->fs_info->delalloc_block_rsv) { + spin_lock(&BTRFS_I(inode)->lock); + if (BTRFS_I(inode)->delalloc_meta_reserved) { + BTRFS_I(inode)->delalloc_meta_reserved = 0; + spin_unlock(&BTRFS_I(inode)->lock); + release = true; + goto migrate; + } + spin_unlock(&BTRFS_I(inode)->lock); + + /* Ok we didn't have space pre-reserved. This shouldn't happen + * too often but it can happen if we do delalloc to an existing + * inode which gets dirtied because of the time update, and then + * isn't touched again until after the transaction commits and + * then we try to write out the data. First try to be nice and + * reserve something strictly for us. If not be a pain and try + * to steal from the delalloc block rsv. + */ + ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes); + if (!ret) + goto out; + + ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes); + if (!ret) + goto out; + + /* + * Ok this is a problem, let's just steal from the global rsv + * since this really shouldn't happen that often. + */ + WARN_ON(1); + ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv, + dst_rsv, num_bytes); + goto out; } +migrate: ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes); + +out: + /* + * Migrate only takes a reservation, it doesn't touch the size of the + * block_rsv. This is to simplify people who don't normally have things + * migrated from their block rsv. If they go to release their + * reservation, that will decrease the size as well, so if migrate + * reduced size we'd end up with a negative size. But for the + * delalloc_meta_reserved stuff we will only know to drop 1 reservation, + * but we could in fact do this reserve/migrate dance several times + * between the time we did the original reservation and we'd clean it + * up. So to take care of this, release the space for the meta + * reservation here. I think it may be time for a documentation page on + * how block rsvs. work. + */ if (!ret) node->bytes_reserved = num_bytes; + if (release) + btrfs_block_rsv_release(root, src_rsv, num_bytes); + return ret; } @@ -1708,7 +1763,8 @@ int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans, goto release_node; } - ret = btrfs_delayed_inode_reserve_metadata(trans, root, delayed_node); + ret = btrfs_delayed_inode_reserve_metadata(trans, root, inode, + delayed_node); if (ret) goto release_node; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 102c176fc29c..62afe5c5694e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1890,31 +1890,32 @@ struct btrfs_root *open_ctree(struct super_block *sb, u64 features; struct btrfs_key location; struct buffer_head *bh; - struct btrfs_root *extent_root = kzalloc(sizeof(struct btrfs_root), - GFP_NOFS); - struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root), - GFP_NOFS); + struct btrfs_super_block *disk_super; struct btrfs_root *tree_root = btrfs_sb(sb); - struct btrfs_fs_info *fs_info = NULL; - struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root), - GFP_NOFS); - struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root), - GFP_NOFS); + struct btrfs_fs_info *fs_info = tree_root->fs_info; + struct btrfs_root *extent_root; + struct btrfs_root *csum_root; + struct btrfs_root *chunk_root; + struct btrfs_root *dev_root; struct btrfs_root *log_tree_root; - int ret; int err = -EINVAL; int num_backups_tried = 0; int backup_index = 0; - struct btrfs_super_block *disk_super; + extent_root = fs_info->extent_root = + kzalloc(sizeof(struct btrfs_root), GFP_NOFS); + csum_root = fs_info->csum_root = + kzalloc(sizeof(struct btrfs_root), GFP_NOFS); + chunk_root = fs_info->chunk_root = + kzalloc(sizeof(struct btrfs_root), GFP_NOFS); + dev_root = fs_info->dev_root = + kzalloc(sizeof(struct btrfs_root), GFP_NOFS); - if (!extent_root || !tree_root || !tree_root->fs_info || - !chunk_root || !dev_root || !csum_root) { + if (!extent_root || !csum_root || !chunk_root || !dev_root) { err = -ENOMEM; goto fail; } - fs_info = tree_root->fs_info; ret = init_srcu_struct(&fs_info->subvol_srcu); if (ret) { @@ -1954,12 +1955,6 @@ struct btrfs_root *open_ctree(struct super_block *sb, mutex_init(&fs_info->reloc_mutex); init_completion(&fs_info->kobj_unregister); - fs_info->tree_root = tree_root; - fs_info->extent_root = extent_root; - fs_info->csum_root = csum_root; - fs_info->chunk_root = chunk_root; - fs_info->dev_root = dev_root; - fs_info->fs_devices = fs_devices; INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots); INIT_LIST_HEAD(&fs_info->space_info); btrfs_mapping_init(&fs_info->mapping_tree); @@ -2465,21 +2460,20 @@ fail_sb_buffer: btrfs_stop_workers(&fs_info->caching_workers); fail_alloc: fail_iput: + btrfs_mapping_tree_free(&fs_info->mapping_tree); + invalidate_inode_pages2(fs_info->btree_inode->i_mapping); iput(fs_info->btree_inode); - - btrfs_close_devices(fs_info->fs_devices); - btrfs_mapping_tree_free(&fs_info->mapping_tree); fail_bdi: bdi_destroy(&fs_info->bdi); fail_srcu: cleanup_srcu_struct(&fs_info->subvol_srcu); fail: + btrfs_close_devices(fs_info->fs_devices); free_fs_info(fs_info); return ERR_PTR(err); recovery_tree_root: - if (!btrfs_test_opt(tree_root, RECOVERY)) goto fail_tree_roots; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 9879bd474632..b232150b5b6b 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3797,16 +3797,16 @@ void btrfs_free_block_rsv(struct btrfs_root *root, kfree(rsv); } -int btrfs_block_rsv_add(struct btrfs_root *root, - struct btrfs_block_rsv *block_rsv, - u64 num_bytes) +static inline int __block_rsv_add(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 num_bytes, int flush) { int ret; if (num_bytes == 0) return 0; - ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 1); + ret = reserve_metadata_bytes(root, block_rsv, num_bytes, flush); if (!ret) { block_rsv_add_bytes(block_rsv, num_bytes, 1); return 0; @@ -3815,22 +3815,18 @@ int btrfs_block_rsv_add(struct btrfs_root *root, return ret; } +int btrfs_block_rsv_add(struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 num_bytes) +{ + return __block_rsv_add(root, block_rsv, num_bytes, 1); +} + int btrfs_block_rsv_add_noflush(struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, u64 num_bytes) { - int ret; - - if (num_bytes == 0) - return 0; - - ret = reserve_metadata_bytes(root, block_rsv, num_bytes, 0); - if (!ret) { - block_rsv_add_bytes(block_rsv, num_bytes, 1); - return 0; - } - - return ret; + return __block_rsv_add(root, block_rsv, num_bytes, 0); } int btrfs_block_rsv_check(struct btrfs_root *root, @@ -4064,23 +4060,30 @@ int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans, */ static unsigned drop_outstanding_extent(struct inode *inode) { + unsigned drop_inode_space = 0; unsigned dropped_extents = 0; BUG_ON(!BTRFS_I(inode)->outstanding_extents); BTRFS_I(inode)->outstanding_extents--; + if (BTRFS_I(inode)->outstanding_extents == 0 && + BTRFS_I(inode)->delalloc_meta_reserved) { + drop_inode_space = 1; + BTRFS_I(inode)->delalloc_meta_reserved = 0; + } + /* * If we have more or the same amount of outsanding extents than we have * reserved then we need to leave the reserved extents count alone. */ if (BTRFS_I(inode)->outstanding_extents >= BTRFS_I(inode)->reserved_extents) - return 0; + return drop_inode_space; dropped_extents = BTRFS_I(inode)->reserved_extents - BTRFS_I(inode)->outstanding_extents; BTRFS_I(inode)->reserved_extents -= dropped_extents; - return dropped_extents; + return dropped_extents + drop_inode_space; } /** @@ -4166,9 +4169,18 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) nr_extents = BTRFS_I(inode)->outstanding_extents - BTRFS_I(inode)->reserved_extents; BTRFS_I(inode)->reserved_extents += nr_extents; + } - to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents); + /* + * Add an item to reserve for updating the inode when we complete the + * delalloc io. + */ + if (!BTRFS_I(inode)->delalloc_meta_reserved) { + nr_extents++; + BTRFS_I(inode)->delalloc_meta_reserved = 1; } + + to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents); to_reserve += calc_csum_metadata_size(inode, num_bytes, 1); spin_unlock(&BTRFS_I(inode)->lock); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 7a15fcfb3e1f..181760f9d2ab 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -537,6 +537,13 @@ static int io_ctl_read_entry(struct io_ctl *io_ctl, struct btrfs_free_space *entry, u8 *type) { struct btrfs_free_space_entry *e; + int ret; + + if (!io_ctl->cur) { + ret = io_ctl_check_crc(io_ctl, io_ctl->index); + if (ret) + return ret; + } e = io_ctl->cur; entry->offset = le64_to_cpu(e->offset); @@ -550,10 +557,7 @@ static int io_ctl_read_entry(struct io_ctl *io_ctl, io_ctl_unmap_page(io_ctl); - if (io_ctl->index >= io_ctl->num_pages) - return 0; - - return io_ctl_check_crc(io_ctl, io_ctl->index); + return 0; } static int io_ctl_read_bitmap(struct io_ctl *io_ctl, @@ -561,9 +565,6 @@ static int io_ctl_read_bitmap(struct io_ctl *io_ctl, { int ret; - if (io_ctl->cur && io_ctl->cur != io_ctl->orig) - io_ctl_unmap_page(io_ctl); - ret = io_ctl_check_crc(io_ctl, io_ctl->index); if (ret) return ret; @@ -699,6 +700,8 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, num_entries--; } + io_ctl_unmap_page(&io_ctl); + /* * We add the bitmaps at the end of the entries in order that * the bitmap entries are added to the cache. diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index 53dcbdf446cd..f8962a957d65 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -398,6 +398,8 @@ int btrfs_save_ino_cache(struct btrfs_root *root, struct btrfs_free_space_ctl *ctl = root->free_ino_ctl; struct btrfs_path *path; struct inode *inode; + struct btrfs_block_rsv *rsv; + u64 num_bytes; u64 alloc_hint = 0; int ret; int prealloc; @@ -421,11 +423,26 @@ int btrfs_save_ino_cache(struct btrfs_root *root, if (!path) return -ENOMEM; + rsv = trans->block_rsv; + trans->block_rsv = &root->fs_info->trans_block_rsv; + + num_bytes = trans->bytes_reserved; + /* + * 1 item for inode item insertion if need + * 3 items for inode item update (in the worst case) + * 1 item for free space object + * 3 items for pre-allocation + */ + trans->bytes_reserved = btrfs_calc_trans_metadata_size(root, 8); + ret = btrfs_block_rsv_add_noflush(root, trans->block_rsv, + trans->bytes_reserved); + if (ret) + goto out; again: inode = lookup_free_ino_inode(root, path); if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) { ret = PTR_ERR(inode); - goto out; + goto out_release; } if (IS_ERR(inode)) { @@ -434,7 +451,7 @@ again: ret = create_free_ino_inode(root, trans, path); if (ret) - goto out; + goto out_release; goto again; } @@ -477,11 +494,14 @@ again: } btrfs_free_reserved_data_space(inode, prealloc); + ret = btrfs_write_out_ino_cache(root, trans, path); out_put: iput(inode); +out_release: + btrfs_block_rsv_release(root, trans->block_rsv, trans->bytes_reserved); out: - if (ret == 0) - ret = btrfs_write_out_ino_cache(root, trans, path); + trans->block_rsv = rsv; + trans->bytes_reserved = num_bytes; btrfs_free_path(path); return ret; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 966ddcc4c63d..116ab67a06df 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -93,6 +93,8 @@ static noinline int cow_file_range(struct inode *inode, struct page *locked_page, u64 start, u64 end, int *page_started, unsigned long *nr_written, int unlock); +static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode); static int btrfs_init_inode_security(struct btrfs_trans_handle *trans, struct inode *inode, struct inode *dir, @@ -1741,7 +1743,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) trans = btrfs_join_transaction(root); BUG_ON(IS_ERR(trans)); trans->block_rsv = &root->fs_info->delalloc_block_rsv; - ret = btrfs_update_inode(trans, root, inode); + ret = btrfs_update_inode_fallback(trans, root, inode); BUG_ON(ret); } goto out; @@ -1791,7 +1793,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent); if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) { - ret = btrfs_update_inode(trans, root, inode); + ret = btrfs_update_inode_fallback(trans, root, inode); BUG_ON(ret); } ret = 0; @@ -2199,6 +2201,9 @@ int btrfs_orphan_cleanup(struct btrfs_root *root) if (ret) goto out; } + /* release the path since we're done with it */ + btrfs_release_path(path); + root->orphan_cleanup_state = ORPHAN_CLEANUP_DONE; if (root->orphan_block_rsv) @@ -2426,7 +2431,7 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, /* * copy everything in the in-memory inode into the btree. */ -noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, +static noinline int btrfs_update_inode_item(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct inode *inode) { struct btrfs_inode_item *inode_item; @@ -2434,21 +2439,6 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, struct extent_buffer *leaf; int ret; - /* - * If the inode is a free space inode, we can deadlock during commit - * if we put it into the delayed code. - * - * The data relocation inode should also be directly updated - * without delay - */ - if (!btrfs_is_free_space_inode(root, inode) - && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) { - ret = btrfs_delayed_update_inode(trans, root, inode); - if (!ret) - btrfs_set_inode_last_trans(trans, inode); - return ret; - } - path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -2477,6 +2467,43 @@ failed: } /* + * copy everything in the in-memory inode into the btree. + */ +noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode) +{ + int ret; + + /* + * If the inode is a free space inode, we can deadlock during commit + * if we put it into the delayed code. + * + * The data relocation inode should also be directly updated + * without delay + */ + if (!btrfs_is_free_space_inode(root, inode) + && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) { + ret = btrfs_delayed_update_inode(trans, root, inode); + if (!ret) + btrfs_set_inode_last_trans(trans, inode); + return ret; + } + + return btrfs_update_inode_item(trans, root, inode); +} + +static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans, + struct btrfs_root *root, struct inode *inode) +{ + int ret; + + ret = btrfs_update_inode(trans, root, inode); + if (ret == -ENOSPC) + return btrfs_update_inode_item(trans, root, inode); + return ret; +} + +/* * unlink helper that gets used here in inode.c and in the tree logging * recovery code. It remove a link in a directory with a given name, and * also drops the back refs in the inode to the directory @@ -5632,7 +5659,7 @@ again: if (test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) { ret = btrfs_ordered_update_i_size(inode, 0, ordered); if (!ret) - err = btrfs_update_inode(trans, root, inode); + err = btrfs_update_inode_fallback(trans, root, inode); goto out; } @@ -5670,7 +5697,7 @@ again: add_pending_csums(trans, inode, ordered->file_offset, &ordered->list); ret = btrfs_ordered_update_i_size(inode, 0, ordered); if (!ret || !test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags)) - btrfs_update_inode(trans, root, inode); + btrfs_update_inode_fallback(trans, root, inode); ret = 0; out_unlock: unlock_extent_cached(&BTRFS_I(inode)->io_tree, ordered->file_offset, @@ -6529,14 +6556,16 @@ end_trans: ret = btrfs_orphan_del(NULL, inode); } - trans->block_rsv = &root->fs_info->trans_block_rsv; - ret = btrfs_update_inode(trans, root, inode); - if (ret && !err) - err = ret; + if (trans) { + trans->block_rsv = &root->fs_info->trans_block_rsv; + ret = btrfs_update_inode(trans, root, inode); + if (ret && !err) + err = ret; - nr = trans->blocks_used; - ret = btrfs_end_transaction_throttle(trans, root); - btrfs_btree_balance_dirty(root, nr); + nr = trans->blocks_used; + ret = btrfs_end_transaction_throttle(trans, root); + btrfs_btree_balance_dirty(root, nr); + } out: btrfs_free_block_rsv(root, rsv); @@ -6605,6 +6634,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->orphan_meta_reserved = 0; ei->dummy_inode = 0; ei->in_defrag = 0; + ei->delalloc_meta_reserved = 0; ei->force_compress = BTRFS_COMPRESS_NONE; ei->delayed_node = NULL; diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 24d654ce7a06..dff29d5e151a 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1174,6 +1174,8 @@ static int clone_backref_node(struct btrfs_trans_handle *trans, list_add_tail(&new_edge->list[UPPER], &new_node->lower); } + } else { + list_add_tail(&new_node->lower, &cache->leaves); } rb_node = tree_insert(&cache->rb_root, new_node->bytenr, diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index ed11d3866afd..f4190f22edfb 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -944,50 +944,18 @@ static int scrub_checksum_super(struct scrub_bio *sbio, void *buffer) static int scrub_submit(struct scrub_dev *sdev) { struct scrub_bio *sbio; - struct bio *bio; - int i; if (sdev->curr == -1) return 0; sbio = sdev->bios[sdev->curr]; - - bio = bio_alloc(GFP_NOFS, sbio->count); - if (!bio) - goto nomem; - - bio->bi_private = sbio; - bio->bi_end_io = scrub_bio_end_io; - bio->bi_bdev = sdev->dev->bdev; - bio->bi_sector = sbio->physical >> 9; - - for (i = 0; i < sbio->count; ++i) { - struct page *page; - int ret; - - page = alloc_page(GFP_NOFS); - if (!page) - goto nomem; - - ret = bio_add_page(bio, page, PAGE_SIZE, 0); - if (!ret) { - __free_page(page); - goto nomem; - } - } - sbio->err = 0; sdev->curr = -1; atomic_inc(&sdev->in_flight); - submit_bio(READ, bio); + submit_bio(READ, sbio->bio); return 0; - -nomem: - scrub_free_bio(bio); - - return -ENOMEM; } static int scrub_page(struct scrub_dev *sdev, u64 logical, u64 len, @@ -995,6 +963,8 @@ static int scrub_page(struct scrub_dev *sdev, u64 logical, u64 len, u8 *csum, int force) { struct scrub_bio *sbio; + struct page *page; + int ret; again: /* @@ -1015,12 +985,22 @@ again: } sbio = sdev->bios[sdev->curr]; if (sbio->count == 0) { + struct bio *bio; + sbio->physical = physical; sbio->logical = logical; + bio = bio_alloc(GFP_NOFS, SCRUB_PAGES_PER_BIO); + if (!bio) + return -ENOMEM; + + bio->bi_private = sbio; + bio->bi_end_io = scrub_bio_end_io; + bio->bi_bdev = sdev->dev->bdev; + bio->bi_sector = sbio->physical >> 9; + sbio->err = 0; + sbio->bio = bio; } else if (sbio->physical + sbio->count * PAGE_SIZE != physical || sbio->logical + sbio->count * PAGE_SIZE != logical) { - int ret; - ret = scrub_submit(sdev); if (ret) return ret; @@ -1030,6 +1010,20 @@ again: sbio->spag[sbio->count].generation = gen; sbio->spag[sbio->count].have_csum = 0; sbio->spag[sbio->count].mirror_num = mirror_num; + + page = alloc_page(GFP_NOFS); + if (!page) + return -ENOMEM; + + ret = bio_add_page(sbio->bio, page, PAGE_SIZE, 0); + if (!ret) { + __free_page(page); + ret = scrub_submit(sdev); + if (ret) + return ret; + goto again; + } + if (csum) { sbio->spag[sbio->count].have_csum = 1; memcpy(sbio->spag[sbio->count].csum, csum, sdev->csum_size); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 57080dffdfc6..8bd9d6d0e07a 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -197,7 +197,7 @@ static match_table_t tokens = { {Opt_subvolrootid, "subvolrootid=%d"}, {Opt_defrag, "autodefrag"}, {Opt_inode_cache, "inode_cache"}, - {Opt_no_space_cache, "no_space_cache"}, + {Opt_no_space_cache, "nospace_cache"}, {Opt_recovery, "recovery"}, {Opt_err, NULL}, }; @@ -448,6 +448,7 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags, token = match_token(p, tokens, args); switch (token) { case Opt_subvol: + kfree(*subvol_name); *subvol_name = match_strdup(&args[0]); break; case Opt_subvolid: @@ -710,7 +711,7 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) if (btrfs_test_opt(root, SPACE_CACHE)) seq_puts(seq, ",space_cache"); else - seq_puts(seq, ",no_space_cache"); + seq_puts(seq, ",nospace_cache"); if (btrfs_test_opt(root, CLEAR_CACHE)) seq_puts(seq, ",clear_cache"); if (btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED)) @@ -890,7 +891,6 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, struct super_block *s; struct dentry *root; struct btrfs_fs_devices *fs_devices = NULL; - struct btrfs_root *tree_root = NULL; struct btrfs_fs_info *fs_info = NULL; fmode_t mode = FMODE_READ; char *subvol_name = NULL; @@ -904,8 +904,10 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, error = btrfs_parse_early_options(data, mode, fs_type, &subvol_name, &subvol_objectid, &subvol_rootid, &fs_devices); - if (error) + if (error) { + kfree(subvol_name); return ERR_PTR(error); + } if (subvol_name) { root = mount_subvol(subvol_name, flags, device_name, data); @@ -917,15 +919,6 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, if (error) return ERR_PTR(error); - error = btrfs_open_devices(fs_devices, mode, fs_type); - if (error) - return ERR_PTR(error); - - if (!(flags & MS_RDONLY) && fs_devices->rw_devices == 0) { - error = -EACCES; - goto error_close_devices; - } - /* * Setup a dummy root and fs_info for test/set super. This is because * we don't actually fill this stuff out until open_ctree, but we need @@ -933,24 +926,36 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, * then open_ctree will properly initialize everything later. */ fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS); - tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS); - if (!fs_info || !tree_root) { + if (!fs_info) + return ERR_PTR(-ENOMEM); + + fs_info->tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS); + if (!fs_info->tree_root) { error = -ENOMEM; - goto error_close_devices; + goto error_fs_info; } - fs_info->tree_root = tree_root; + fs_info->tree_root->fs_info = fs_info; fs_info->fs_devices = fs_devices; - tree_root->fs_info = fs_info; fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS); fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_NOFS); if (!fs_info->super_copy || !fs_info->super_for_commit) { error = -ENOMEM; + goto error_fs_info; + } + + error = btrfs_open_devices(fs_devices, mode, fs_type); + if (error) + goto error_fs_info; + + if (!(flags & MS_RDONLY) && fs_devices->rw_devices == 0) { + error = -EACCES; goto error_close_devices; } bdev = fs_devices->latest_bdev; - s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root); + s = sget(fs_type, btrfs_test_super, btrfs_set_super, + fs_info->tree_root); if (IS_ERR(s)) { error = PTR_ERR(s); goto error_close_devices; @@ -959,12 +964,12 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, if (s->s_root) { if ((flags ^ s->s_flags) & MS_RDONLY) { deactivate_locked_super(s); - return ERR_PTR(-EBUSY); + error = -EBUSY; + goto error_close_devices; } btrfs_close_devices(fs_devices); free_fs_info(fs_info); - kfree(tree_root); } else { char b[BDEVNAME_SIZE]; @@ -991,8 +996,8 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, error_close_devices: btrfs_close_devices(fs_devices); +error_fs_info: free_fs_info(fs_info); - kfree(tree_root); return ERR_PTR(error); } diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 960835eaf4da..6a0574e923bc 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -882,8 +882,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_reloc_pre_snapshot(trans, pending, &to_reserve); if (to_reserve > 0) { - ret = btrfs_block_rsv_add(root, &pending->block_rsv, - to_reserve); + ret = btrfs_block_rsv_add_noflush(root, &pending->block_rsv, + to_reserve); if (ret) { pending->error = ret; goto fail; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index f8e2943101a1..c37433d3cd82 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -999,7 +999,7 @@ static int btrfs_free_dev_extent(struct btrfs_trans_handle *trans, key.objectid = device->devid; key.offset = start; key.type = BTRFS_DEV_EXTENT_KEY; - +again: ret = btrfs_search_slot(trans, root, &key, path, -1, 1); if (ret > 0) { ret = btrfs_previous_item(root, path, key.objectid, @@ -1012,6 +1012,9 @@ static int btrfs_free_dev_extent(struct btrfs_trans_handle *trans, struct btrfs_dev_extent); BUG_ON(found_key.offset > start || found_key.offset + btrfs_dev_extent_length(leaf, extent) < start); + key = found_key; + btrfs_release_path(path); + goto again; } else if (ret == 0) { leaf = path->nodes[0]; extent = btrfs_item_ptr(leaf, path->slots[0], diff --git a/fs/cifs/file.c b/fs/cifs/file.c index c1f063cd1b0c..cf0b1539b321 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -645,20 +645,20 @@ int cifs_closedir(struct inode *inode, struct file *file) } static struct cifsLockInfo * -cifs_lock_init(__u64 len, __u64 offset, __u8 type, __u16 netfid) +cifs_lock_init(__u64 offset, __u64 length, __u8 type, __u16 netfid) { - struct cifsLockInfo *li = + struct cifsLockInfo *lock = kmalloc(sizeof(struct cifsLockInfo), GFP_KERNEL); - if (!li) - return li; - li->netfid = netfid; - li->offset = offset; - li->length = len; - li->type = type; - li->pid = current->tgid; - INIT_LIST_HEAD(&li->blist); - init_waitqueue_head(&li->block_q); - return li; + if (!lock) + return lock; + lock->offset = offset; + lock->length = length; + lock->type = type; + lock->netfid = netfid; + lock->pid = current->tgid; + INIT_LIST_HEAD(&lock->blist); + init_waitqueue_head(&lock->block_q); + return lock; } static void @@ -672,7 +672,7 @@ cifs_del_lock_waiters(struct cifsLockInfo *lock) } static bool -cifs_find_lock_conflict(struct cifsInodeInfo *cinode, __u64 offset, +__cifs_find_lock_conflict(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, __u8 type, __u16 netfid, struct cifsLockInfo **conf_lock) { @@ -694,6 +694,14 @@ cifs_find_lock_conflict(struct cifsInodeInfo *cinode, __u64 offset, return false; } +static bool +cifs_find_lock_conflict(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock, + struct cifsLockInfo **conf_lock) +{ + return __cifs_find_lock_conflict(cinode, lock->offset, lock->length, + lock->type, lock->netfid, conf_lock); +} + static int cifs_lock_test(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, __u8 type, __u16 netfid, struct file_lock *flock) @@ -704,8 +712,8 @@ cifs_lock_test(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, mutex_lock(&cinode->lock_mutex); - exist = cifs_find_lock_conflict(cinode, offset, length, type, netfid, - &conf_lock); + exist = __cifs_find_lock_conflict(cinode, offset, length, type, netfid, + &conf_lock); if (exist) { flock->fl_start = conf_lock->offset; flock->fl_end = conf_lock->offset + conf_lock->length - 1; @@ -723,40 +731,27 @@ cifs_lock_test(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, return rc; } -static int -cifs_lock_add(struct cifsInodeInfo *cinode, __u64 len, __u64 offset, - __u8 type, __u16 netfid) +static void +cifs_lock_add(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock) { - struct cifsLockInfo *li; - - li = cifs_lock_init(len, offset, type, netfid); - if (!li) - return -ENOMEM; - mutex_lock(&cinode->lock_mutex); - list_add_tail(&li->llist, &cinode->llist); + list_add_tail(&lock->llist, &cinode->llist); mutex_unlock(&cinode->lock_mutex); - return 0; } static int -cifs_lock_add_if(struct cifsInodeInfo *cinode, __u64 offset, __u64 length, - __u8 type, __u16 netfid, bool wait) +cifs_lock_add_if(struct cifsInodeInfo *cinode, struct cifsLockInfo *lock, + bool wait) { - struct cifsLockInfo *lock, *conf_lock; + struct cifsLockInfo *conf_lock; bool exist; int rc = 0; - lock = cifs_lock_init(length, offset, type, netfid); - if (!lock) - return -ENOMEM; - try_again: exist = false; mutex_lock(&cinode->lock_mutex); - exist = cifs_find_lock_conflict(cinode, offset, length, type, netfid, - &conf_lock); + exist = cifs_find_lock_conflict(cinode, lock, &conf_lock); if (!exist && cinode->can_cache_brlcks) { list_add_tail(&lock->llist, &cinode->llist); mutex_unlock(&cinode->lock_mutex); @@ -775,13 +770,10 @@ try_again: (lock->blist.next == &lock->blist)); if (!rc) goto try_again; - else { - mutex_lock(&cinode->lock_mutex); - list_del_init(&lock->blist); - } + mutex_lock(&cinode->lock_mutex); + list_del_init(&lock->blist); } - kfree(lock); mutex_unlock(&cinode->lock_mutex); return rc; } @@ -933,7 +925,7 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile) else type = CIFS_WRLCK; - lck = cifs_lock_init(length, flock->fl_start, type, + lck = cifs_lock_init(flock->fl_start, length, type, cfile->netfid); if (!lck) { rc = -ENOMEM; @@ -1070,14 +1062,12 @@ cifs_getlk(struct file *file, struct file_lock *flock, __u8 type, if (rc != 0) cERROR(1, "Error unlocking previously locked " "range %d during test of lock", rc); - rc = 0; - return rc; + return 0; } if (type & LOCKING_ANDX_SHARED_LOCK) { flock->fl_type = F_WRLCK; - rc = 0; - return rc; + return 0; } rc = CIFSSMBLock(xid, tcon, netfid, current->tgid, length, @@ -1095,8 +1085,7 @@ cifs_getlk(struct file *file, struct file_lock *flock, __u8 type, } else flock->fl_type = F_WRLCK; - rc = 0; - return rc; + return 0; } static void @@ -1254,20 +1243,26 @@ cifs_setlk(struct file *file, struct file_lock *flock, __u8 type, } if (lock) { - rc = cifs_lock_add_if(cinode, flock->fl_start, length, - type, netfid, wait_flag); + struct cifsLockInfo *lock; + + lock = cifs_lock_init(flock->fl_start, length, type, netfid); + if (!lock) + return -ENOMEM; + + rc = cifs_lock_add_if(cinode, lock, wait_flag); if (rc < 0) - return rc; - else if (!rc) + kfree(lock); + if (rc <= 0) goto out; rc = CIFSSMBLock(xid, tcon, netfid, current->tgid, length, flock->fl_start, 0, 1, type, wait_flag, 0); - if (rc == 0) { - /* For Windows locks we must store them. */ - rc = cifs_lock_add(cinode, length, flock->fl_start, - type, netfid); + if (rc) { + kfree(lock); + goto out; } + + cifs_lock_add(cinode, lock); } else if (unlock) rc = cifs_unlock_range(cfile, flock, xid); diff --git a/fs/dcache.c b/fs/dcache.c index 274f13e2f094..a901c6901bce 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -546,9 +546,11 @@ int d_invalidate(struct dentry * dentry) * would make it unreachable from the root, * we might still populate it if it was a * working directory or similar). + * We also need to leave mountpoints alone, + * directory or not. */ - if (dentry->d_count > 1) { - if (dentry->d_inode && S_ISDIR(dentry->d_inode->i_mode)) { + if (dentry->d_count > 1 && dentry->d_inode) { + if (S_ISDIR(dentry->d_inode->i_mode) || d_mountpoint(dentry)) { spin_unlock(&dentry->d_lock); return -EBUSY; } diff --git a/fs/hfs/trans.c b/fs/hfs/trans.c index e673a88b8ae7..b1ce4c7ad3fb 100644 --- a/fs/hfs/trans.c +++ b/fs/hfs/trans.c @@ -40,6 +40,8 @@ int hfs_mac2asc(struct super_block *sb, char *out, const struct hfs_name *in) src = in->name; srclen = in->len; + if (srclen > HFS_NAMELEN) + srclen = HFS_NAMELEN; dst = out; dstlen = HFS_MAX_NAMELEN; if (nls_io) { diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c index de4247021d25..5b6c9d1a2fb9 100644 --- a/fs/jffs2/compr.c +++ b/fs/jffs2/compr.c @@ -53,6 +53,78 @@ static int jffs2_is_best_compression(struct jffs2_compressor *this, return 0; } +/* + * jffs2_selected_compress: + * @compr: Explicit compression type to use (ie, JFFS2_COMPR_ZLIB). + * If 0, just take the first available compression mode. + * @data_in: Pointer to uncompressed data + * @cpage_out: Pointer to returned pointer to buffer for compressed data + * @datalen: On entry, holds the amount of data available for compression. + * On exit, expected to hold the amount of data actually compressed. + * @cdatalen: On entry, holds the amount of space available for compressed + * data. On exit, expected to hold the actual size of the compressed + * data. + * + * Returns: the compression type used. Zero is used to show that the data + * could not be compressed; probably because we couldn't find the requested + * compression mode. + */ +static int jffs2_selected_compress(u8 compr, unsigned char *data_in, + unsigned char **cpage_out, u32 *datalen, u32 *cdatalen) +{ + struct jffs2_compressor *this; + int err, ret = JFFS2_COMPR_NONE; + uint32_t orig_slen, orig_dlen; + char *output_buf; + + output_buf = kmalloc(*cdatalen, GFP_KERNEL); + if (!output_buf) { + printk(KERN_WARNING "JFFS2: No memory for compressor allocation. Compression failed.\n"); + return ret; + } + orig_slen = *datalen; + orig_dlen = *cdatalen; + spin_lock(&jffs2_compressor_list_lock); + list_for_each_entry(this, &jffs2_compressor_list, list) { + /* Skip decompress-only and disabled modules */ + if (!this->compress || this->disabled) + continue; + + /* Skip if not the desired compression type */ + if (compr && (compr != this->compr)) + continue; + + /* + * Either compression type was unspecified, or we found our + * compressor; either way, we're good to go. + */ + this->usecount++; + spin_unlock(&jffs2_compressor_list_lock); + + *datalen = orig_slen; + *cdatalen = orig_dlen; + err = this->compress(data_in, output_buf, datalen, cdatalen); + + spin_lock(&jffs2_compressor_list_lock); + this->usecount--; + if (!err) { + /* Success */ + ret = this->compr; + this->stat_compr_blocks++; + this->stat_compr_orig_size += *datalen; + this->stat_compr_new_size += *cdatalen; + break; + } + } + spin_unlock(&jffs2_compressor_list_lock); + if (ret == JFFS2_COMPR_NONE) + kfree(output_buf); + else + *cpage_out = output_buf; + + return ret; +} + /* jffs2_compress: * @data_in: Pointer to uncompressed data * @cpage_out: Pointer to returned pointer to buffer for compressed data @@ -76,47 +148,23 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint32_t *datalen, uint32_t *cdatalen) { int ret = JFFS2_COMPR_NONE; - int compr_ret; + int mode, compr_ret; struct jffs2_compressor *this, *best=NULL; unsigned char *output_buf = NULL, *tmp_buf; uint32_t orig_slen, orig_dlen; uint32_t best_slen=0, best_dlen=0; - switch (jffs2_compression_mode) { + if (c->mount_opts.override_compr) + mode = c->mount_opts.compr; + else + mode = jffs2_compression_mode; + + switch (mode) { case JFFS2_COMPR_MODE_NONE: break; case JFFS2_COMPR_MODE_PRIORITY: - output_buf = kmalloc(*cdatalen,GFP_KERNEL); - if (!output_buf) { - printk(KERN_WARNING "JFFS2: No memory for compressor allocation. Compression failed.\n"); - goto out; - } - orig_slen = *datalen; - orig_dlen = *cdatalen; - spin_lock(&jffs2_compressor_list_lock); - list_for_each_entry(this, &jffs2_compressor_list, list) { - /* Skip decompress-only backwards-compatibility and disabled modules */ - if ((!this->compress)||(this->disabled)) - continue; - - this->usecount++; - spin_unlock(&jffs2_compressor_list_lock); - *datalen = orig_slen; - *cdatalen = orig_dlen; - compr_ret = this->compress(data_in, output_buf, datalen, cdatalen); - spin_lock(&jffs2_compressor_list_lock); - this->usecount--; - if (!compr_ret) { - ret = this->compr; - this->stat_compr_blocks++; - this->stat_compr_orig_size += *datalen; - this->stat_compr_new_size += *cdatalen; - break; - } - } - spin_unlock(&jffs2_compressor_list_lock); - if (ret == JFFS2_COMPR_NONE) - kfree(output_buf); + ret = jffs2_selected_compress(0, data_in, cpage_out, datalen, + cdatalen); break; case JFFS2_COMPR_MODE_SIZE: case JFFS2_COMPR_MODE_FAVOURLZO: @@ -174,22 +222,28 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, best->stat_compr_orig_size += best_slen; best->stat_compr_new_size += best_dlen; ret = best->compr; + *cpage_out = output_buf; } spin_unlock(&jffs2_compressor_list_lock); break; + case JFFS2_COMPR_MODE_FORCELZO: + ret = jffs2_selected_compress(JFFS2_COMPR_LZO, data_in, + cpage_out, datalen, cdatalen); + break; + case JFFS2_COMPR_MODE_FORCEZLIB: + ret = jffs2_selected_compress(JFFS2_COMPR_ZLIB, data_in, + cpage_out, datalen, cdatalen); + break; default: printk(KERN_ERR "JFFS2: unknown compression mode.\n"); } - out: + if (ret == JFFS2_COMPR_NONE) { *cpage_out = data_in; *datalen = *cdatalen; none_stat_compr_blocks++; none_stat_compr_size += *datalen; } - else { - *cpage_out = output_buf; - } return ret; } diff --git a/fs/jffs2/compr.h b/fs/jffs2/compr.h index 13bb7597ab39..5e91d578f4ed 100644 --- a/fs/jffs2/compr.h +++ b/fs/jffs2/compr.h @@ -40,6 +40,8 @@ #define JFFS2_COMPR_MODE_PRIORITY 1 #define JFFS2_COMPR_MODE_SIZE 2 #define JFFS2_COMPR_MODE_FAVOURLZO 3 +#define JFFS2_COMPR_MODE_FORCELZO 4 +#define JFFS2_COMPR_MODE_FORCEZLIB 5 #define FAVOUR_LZO_PERCENT 80 diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 7286e44ac665..4b8afe39a87f 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -379,7 +379,7 @@ void jffs2_dirty_inode(struct inode *inode, int flags) jffs2_do_setattr(inode, &iattr); } -int jffs2_remount_fs (struct super_block *sb, int *flags, char *data) +int jffs2_do_remount_fs(struct super_block *sb, int *flags, char *data) { struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); diff --git a/fs/jffs2/jffs2_fs_sb.h b/fs/jffs2/jffs2_fs_sb.h index 0bc6a6c80a56..55a0c1dceadf 100644 --- a/fs/jffs2/jffs2_fs_sb.h +++ b/fs/jffs2/jffs2_fs_sb.h @@ -29,6 +29,11 @@ struct jffs2_inodirty; +struct jffs2_mount_opts { + bool override_compr; + unsigned int compr; +}; + /* A struct for the overall file system control. Pointers to jffs2_sb_info structs are named `c' in the source code. Nee jffs_control @@ -126,6 +131,7 @@ struct jffs2_sb_info { #endif struct jffs2_summary *summary; /* Summary information */ + struct jffs2_mount_opts mount_opts; #ifdef CONFIG_JFFS2_FS_XATTR #define XATTRINDEX_HASHSIZE (57) diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index 6c1755c59c0f..ab65ee3ec858 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -176,7 +176,7 @@ void jffs2_dirty_inode(struct inode *inode, int flags); struct inode *jffs2_new_inode (struct inode *dir_i, umode_t mode, struct jffs2_raw_inode *ri); int jffs2_statfs (struct dentry *, struct kstatfs *); -int jffs2_remount_fs (struct super_block *, int *, char *); +int jffs2_do_remount_fs(struct super_block *, int *, char *); int jffs2_do_fill_super(struct super_block *sb, void *data, int silent); void jffs2_gc_release_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f); diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 8d8cd3419d02..28107ca136e4 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -275,9 +275,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) else c->mtd->unpoint(c->mtd, 0, c->mtd->size); #endif - if (s) - kfree(s); - + kfree(s); return ret; } diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 853b8e300084..e7e974454115 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -17,11 +17,13 @@ #include <linux/fs.h> #include <linux/err.h> #include <linux/mount.h> +#include <linux/parser.h> #include <linux/jffs2.h> #include <linux/pagemap.h> #include <linux/mtd/super.h> #include <linux/ctype.h> #include <linux/namei.h> +#include <linux/seq_file.h> #include <linux/exportfs.h> #include "compr.h" #include "nodelist.h" @@ -75,6 +77,37 @@ static void jffs2_write_super(struct super_block *sb) unlock_super(sb); } +static const char *jffs2_compr_name(unsigned int compr) +{ + switch (compr) { + case JFFS2_COMPR_MODE_NONE: + return "none"; +#ifdef CONFIG_JFFS2_LZO + case JFFS2_COMPR_MODE_FORCELZO: + return "lzo"; +#endif +#ifdef CONFIG_JFFS2_ZLIB + case JFFS2_COMPR_MODE_FORCEZLIB: + return "zlib"; +#endif + default: + /* should never happen; programmer error */ + WARN_ON(1); + return ""; + } +} + +static int jffs2_show_options(struct seq_file *s, struct vfsmount *mnt) +{ + struct jffs2_sb_info *c = JFFS2_SB_INFO(mnt->mnt_sb); + struct jffs2_mount_opts *opts = &c->mount_opts; + + if (opts->override_compr) + seq_printf(s, ",compr=%s", jffs2_compr_name(opts->compr)); + + return 0; +} + static int jffs2_sync_fs(struct super_block *sb, int wait) { struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); @@ -133,6 +166,85 @@ static const struct export_operations jffs2_export_ops = { .fh_to_parent = jffs2_fh_to_parent, }; +/* + * JFFS2 mount options. + * + * Opt_override_compr: override default compressor + * Opt_err: just end of array marker + */ +enum { + Opt_override_compr, + Opt_err, +}; + +static const match_table_t tokens = { + {Opt_override_compr, "compr=%s"}, + {Opt_err, NULL}, +}; + +static int jffs2_parse_options(struct jffs2_sb_info *c, char *data) +{ + substring_t args[MAX_OPT_ARGS]; + char *p, *name; + + if (!data) + return 0; + + while ((p = strsep(&data, ","))) { + int token; + + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_override_compr: + name = match_strdup(&args[0]); + + if (!name) + return -ENOMEM; + if (!strcmp(name, "none")) + c->mount_opts.compr = JFFS2_COMPR_MODE_NONE; +#ifdef CONFIG_JFFS2_LZO + else if (!strcmp(name, "lzo")) + c->mount_opts.compr = JFFS2_COMPR_MODE_FORCELZO; +#endif +#ifdef CONFIG_JFFS2_ZLIB + else if (!strcmp(name, "zlib")) + c->mount_opts.compr = + JFFS2_COMPR_MODE_FORCEZLIB; +#endif + else { + printk(KERN_ERR "JFFS2 Error: unknown compressor \"%s\"", + name); + kfree(name); + return -EINVAL; + } + kfree(name); + c->mount_opts.override_compr = true; + break; + default: + printk(KERN_ERR "JFFS2 Error: unrecognized mount option '%s' or missing value\n", + p); + return -EINVAL; + } + } + + return 0; +} + +static int jffs2_remount_fs(struct super_block *sb, int *flags, char *data) +{ + struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); + int err; + + err = jffs2_parse_options(c, data); + if (err) + return -EINVAL; + + return jffs2_do_remount_fs(sb, flags, data); +} + static const struct super_operations jffs2_super_operations = { .alloc_inode = jffs2_alloc_inode, @@ -143,6 +255,7 @@ static const struct super_operations jffs2_super_operations = .remount_fs = jffs2_remount_fs, .evict_inode = jffs2_evict_inode, .dirty_inode = jffs2_dirty_inode, + .show_options = jffs2_show_options, .sync_fs = jffs2_sync_fs, }; @@ -166,6 +279,12 @@ static int jffs2_fill_super(struct super_block *sb, void *data, int silent) c->os_priv = sb; sb->s_fs_info = c; + ret = jffs2_parse_options(c, data); + if (ret) { + kfree(c); + return -EINVAL; + } + /* Initialize JFFS2 superblock locks, the further initialization will * be done later */ mutex_init(&c->alloc_sem); diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 4515bea0268f..b09e51d2f81f 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -578,8 +578,7 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) if (!jffs2_is_writebuffered(c)) return 0; - if (mutex_trylock(&c->alloc_sem)) { - mutex_unlock(&c->alloc_sem); + if (!mutex_is_locked(&c->alloc_sem)) { printk(KERN_CRIT "jffs2_flush_wbuf() called with alloc_sem not locked!\n"); BUG(); } @@ -1026,7 +1025,7 @@ int jffs2_check_oob_empty(struct jffs2_sb_info *c, int cmlen = min_t(int, c->oobavail, OOB_CM_SIZE); struct mtd_oob_ops ops; - ops.mode = MTD_OOB_AUTO; + ops.mode = MTD_OPS_AUTO_OOB; ops.ooblen = NR_OOB_SCAN_PAGES * c->oobavail; ops.oobbuf = c->oobbuf; ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; @@ -1069,7 +1068,7 @@ int jffs2_check_nand_cleanmarker(struct jffs2_sb_info *c, struct mtd_oob_ops ops; int ret, cmlen = min_t(int, c->oobavail, OOB_CM_SIZE); - ops.mode = MTD_OOB_AUTO; + ops.mode = MTD_OPS_AUTO_OOB; ops.ooblen = cmlen; ops.oobbuf = c->oobbuf; ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; @@ -1095,7 +1094,7 @@ int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, struct mtd_oob_ops ops; int cmlen = min_t(int, c->oobavail, OOB_CM_SIZE); - ops.mode = MTD_OOB_AUTO; + ops.mode = MTD_OPS_AUTO_OOB; ops.ooblen = cmlen; ops.oobbuf = (uint8_t *)&oob_cleanmarker; ops.len = ops.ooboffs = ops.retlen = ops.oobretlen = 0; diff --git a/fs/namei.c b/fs/namei.c index ac6d214da827..5008f01787f5 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -852,7 +852,7 @@ static int follow_managed(struct path *path, unsigned flags) mntput(path->mnt); if (ret == -EISDIR) ret = 0; - return ret; + return ret < 0 ? ret : need_mntput; } int follow_down_one(struct path *path) @@ -900,6 +900,7 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, break; path->mnt = mounted; path->dentry = mounted->mnt_root; + nd->flags |= LOOKUP_JUMPED; nd->seq = read_seqcount_begin(&path->dentry->d_seq); /* * Update the inode too. We don't need to re-check the @@ -1213,6 +1214,8 @@ retry: path_put_conditional(path, nd); return err; } + if (err) + nd->flags |= LOOKUP_JUMPED; *inode = path->dentry->d_inode; return 0; } @@ -2146,6 +2149,10 @@ static struct file *do_last(struct nameidata *nd, struct path *path, } /* create side of things */ + /* + * This will *only* deal with leaving RCU mode - LOOKUP_JUMPED has been + * cleared when we got to the last component we are about to look up + */ error = complete_walk(nd); if (error) return ERR_PTR(error); @@ -2214,6 +2221,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, if (error < 0) goto exit_dput; + if (error) + nd->flags |= LOOKUP_JUMPED; + error = -ENOENT; if (!path->dentry->d_inode) goto exit_dput; @@ -2223,6 +2233,10 @@ static struct file *do_last(struct nameidata *nd, struct path *path, path_to_nameidata(path, nd); nd->inode = path->dentry->d_inode; + /* Why this, you ask? _Now_ we might have grown LOOKUP_JUMPED... */ + error = complete_walk(nd); + if (error) + goto exit; error = -EISDIR; if (S_ISDIR(nd->inode->i_mode)) goto exit; diff --git a/fs/proc/base.c b/fs/proc/base.c index 2db1bd3173b2..851ba3dcdc29 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1652,46 +1652,12 @@ out: return error; } -static int proc_pid_fd_link_getattr(struct vfsmount *mnt, struct dentry *dentry, - struct kstat *stat) -{ - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); - int rc; - - if (task == NULL) - return -ESRCH; - - rc = -EACCES; - if (lock_trace(task)) - goto out_task; - - generic_fillattr(inode, stat); - unlock_trace(task); - rc = 0; -out_task: - put_task_struct(task); - return rc; -} - static const struct inode_operations proc_pid_link_inode_operations = { .readlink = proc_pid_readlink, .follow_link = proc_pid_follow_link, .setattr = proc_setattr, }; -static const struct inode_operations proc_fdinfo_link_inode_operations = { - .setattr = proc_setattr, - .getattr = proc_pid_fd_link_getattr, -}; - -static const struct inode_operations proc_fd_link_inode_operations = { - .readlink = proc_pid_readlink, - .follow_link = proc_pid_follow_link, - .setattr = proc_setattr, - .getattr = proc_pid_fd_link_getattr, -}; - /* building an inode */ @@ -1923,61 +1889,49 @@ out: static int proc_fd_info(struct inode *inode, struct path *path, char *info) { - struct task_struct *task; - struct files_struct *files; + struct task_struct *task = get_proc_task(inode); + struct files_struct *files = NULL; struct file *file; int fd = proc_fd(inode); - int rc; - - task = get_proc_task(inode); - if (!task) - return -ENOENT; - - rc = -EACCES; - if (lock_trace(task)) - goto out_task; - - rc = -ENOENT; - files = get_files_struct(task); - if (files == NULL) - goto out_unlock; - /* - * We are not taking a ref to the file structure, so we must - * hold ->file_lock. - */ - spin_lock(&files->file_lock); - file = fcheck_files(files, fd); - if (file) { - unsigned int f_flags; - struct fdtable *fdt; - - fdt = files_fdtable(files); - f_flags = file->f_flags & ~O_CLOEXEC; - if (FD_ISSET(fd, fdt->close_on_exec)) - f_flags |= O_CLOEXEC; - - if (path) { - *path = file->f_path; - path_get(&file->f_path); + if (task) { + files = get_files_struct(task); + put_task_struct(task); + } + if (files) { + /* + * We are not taking a ref to the file structure, so we must + * hold ->file_lock. + */ + spin_lock(&files->file_lock); + file = fcheck_files(files, fd); + if (file) { + unsigned int f_flags; + struct fdtable *fdt; + + fdt = files_fdtable(files); + f_flags = file->f_flags & ~O_CLOEXEC; + if (FD_ISSET(fd, fdt->close_on_exec)) + f_flags |= O_CLOEXEC; + + if (path) { + *path = file->f_path; + path_get(&file->f_path); + } + if (info) + snprintf(info, PROC_FDINFO_MAX, + "pos:\t%lli\n" + "flags:\t0%o\n", + (long long) file->f_pos, + f_flags); + spin_unlock(&files->file_lock); + put_files_struct(files); + return 0; } - if (info) - snprintf(info, PROC_FDINFO_MAX, - "pos:\t%lli\n" - "flags:\t0%o\n", - (long long) file->f_pos, - f_flags); - rc = 0; - } else - rc = -ENOENT; - spin_unlock(&files->file_lock); - put_files_struct(files); - -out_unlock: - unlock_trace(task); -out_task: - put_task_struct(task); - return rc; + spin_unlock(&files->file_lock); + put_files_struct(files); + } + return -ENOENT; } static int proc_fd_link(struct inode *inode, struct path *path) @@ -2072,7 +2026,7 @@ static struct dentry *proc_fd_instantiate(struct inode *dir, spin_unlock(&files->file_lock); put_files_struct(files); - inode->i_op = &proc_fd_link_inode_operations; + inode->i_op = &proc_pid_link_inode_operations; inode->i_size = 64; ei->op.proc_get_link = proc_fd_link; d_set_d_op(dentry, &tid_fd_dentry_operations); @@ -2104,12 +2058,7 @@ static struct dentry *proc_lookupfd_common(struct inode *dir, if (fd == ~0U) goto out; - result = ERR_PTR(-EACCES); - if (lock_trace(task)) - goto out; - result = instantiate(dir, dentry, task, &fd); - unlock_trace(task); out: put_task_struct(task); out_no_task: @@ -2129,28 +2078,23 @@ static int proc_readfd_common(struct file * filp, void * dirent, retval = -ENOENT; if (!p) goto out_no_task; - - retval = -EACCES; - if (lock_trace(p)) - goto out; - retval = 0; fd = filp->f_pos; switch (fd) { case 0: if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) - goto out_unlock; + goto out; filp->f_pos++; case 1: ino = parent_ino(dentry); if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) - goto out_unlock; + goto out; filp->f_pos++; default: files = get_files_struct(p); if (!files) - goto out_unlock; + goto out; rcu_read_lock(); for (fd = filp->f_pos-2; fd < files_fdtable(files)->max_fds; @@ -2174,9 +2118,6 @@ static int proc_readfd_common(struct file * filp, void * dirent, rcu_read_unlock(); put_files_struct(files); } - -out_unlock: - unlock_trace(p); out: put_task_struct(p); out_no_task: @@ -2254,7 +2195,6 @@ static struct dentry *proc_fdinfo_instantiate(struct inode *dir, ei->fd = fd; inode->i_mode = S_IFREG | S_IRUSR; inode->i_fop = &proc_fdinfo_file_operations; - inode->i_op = &proc_fdinfo_link_inode_operations; d_set_d_op(dentry, &tid_fd_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index eef109a1a927..b09ba2dd8b62 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -870,6 +870,22 @@ void dbg_dump_lpt_info(struct ubifs_info *c) spin_unlock(&dbg_lock); } +void dbg_dump_sleb(const struct ubifs_info *c, + const struct ubifs_scan_leb *sleb, int offs) +{ + struct ubifs_scan_node *snod; + + printk(KERN_DEBUG "(pid %d) start dumping scanned data from LEB %d:%d\n", + current->pid, sleb->lnum, offs); + + list_for_each_entry(snod, &sleb->nodes, list) { + cond_resched(); + printk(KERN_DEBUG "Dumping node at LEB %d:%d len %d\n", sleb->lnum, + snod->offs, snod->len); + dbg_dump_node(c, snod->node); + } +} + void dbg_dump_leb(const struct ubifs_info *c, int lnum) { struct ubifs_scan_leb *sleb; diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index feb361e252ac..8d9c46810189 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -269,6 +269,8 @@ void dbg_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp); void dbg_dump_lprops(struct ubifs_info *c); void dbg_dump_lpt_info(struct ubifs_info *c); void dbg_dump_leb(const struct ubifs_info *c, int lnum); +void dbg_dump_sleb(const struct ubifs_info *c, + const struct ubifs_scan_leb *sleb, int offs); void dbg_dump_znode(const struct ubifs_info *c, const struct ubifs_znode *znode); void dbg_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat); @@ -387,6 +389,9 @@ static inline void dbg_dump_lpt_info(struct ubifs_info *c) { return; } static inline void dbg_dump_leb(const struct ubifs_info *c, int lnum) { return; } static inline void +dbg_dump_sleb(const struct ubifs_info *c, + const struct ubifs_scan_leb *sleb, int offs) { return; } +static inline void dbg_dump_znode(const struct ubifs_info *c, const struct ubifs_znode *znode) { return; } static inline void dbg_dump_heap(struct ubifs_info *c, diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c index af02790d9328..ee4f43f4bb99 100644 --- a/fs/ubifs/recovery.c +++ b/fs/ubifs/recovery.c @@ -983,7 +983,7 @@ int ubifs_recover_inl_heads(struct ubifs_info *c, void *sbuf) } /** - * clean_an_unclean_leb - read and write a LEB to remove corruption. + * clean_an_unclean_leb - read and write a LEB to remove corruption. * @c: UBIFS file-system description object * @ucleb: unclean LEB information * @sbuf: LEB-sized buffer to use diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c index 93d938ad3d2a..6094c5a5d7a8 100644 --- a/fs/ubifs/sb.c +++ b/fs/ubifs/sb.c @@ -247,7 +247,7 @@ static int create_default_filesystem(struct ubifs_info *c) mst->total_dirty = cpu_to_le64(tmp64); /* The indexing LEB does not contribute to dark space */ - tmp64 = (c->main_lebs - 1) * c->dark_wm; + tmp64 = ((long long)(c->main_lebs - 1) * c->dark_wm); mst->total_dark = cpu_to_le64(tmp64); mst->total_used = cpu_to_le64(UBIFS_INO_NODE_SZ); diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 33b13310ee0c..574d4ee9b625 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -189,7 +189,7 @@ xfs_end_io( int error = 0; if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { - error = -EIO; + ioend->io_error = -EIO; goto done; } if (ioend->io_error) diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index 1a3513881bce..eac97ef81e2a 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -656,7 +656,7 @@ xfs_buf_item_committing( /* * This is the ops vector shared by all buf log items. */ -static struct xfs_item_ops xfs_buf_item_ops = { +static const struct xfs_item_ops xfs_buf_item_ops = { .iop_size = xfs_buf_item_size, .iop_format = xfs_buf_item_format, .iop_pin = xfs_buf_item_pin, diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c index bb3f71d236d2..0dee0b71029d 100644 --- a/fs/xfs/xfs_dquot_item.c +++ b/fs/xfs/xfs_dquot_item.c @@ -295,7 +295,7 @@ xfs_qm_dquot_logitem_committing( /* * This is the ops vector for dquots */ -static struct xfs_item_ops xfs_dquot_item_ops = { +static const struct xfs_item_ops xfs_dquot_item_ops = { .iop_size = xfs_qm_dquot_logitem_size, .iop_format = xfs_qm_dquot_logitem_format, .iop_pin = xfs_qm_dquot_logitem_pin, @@ -483,7 +483,7 @@ xfs_qm_qoff_logitem_committing( { } -static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = { +static const struct xfs_item_ops xfs_qm_qoffend_logitem_ops = { .iop_size = xfs_qm_qoff_logitem_size, .iop_format = xfs_qm_qoff_logitem_format, .iop_pin = xfs_qm_qoff_logitem_pin, @@ -498,7 +498,7 @@ static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = { /* * This is the ops vector shared by all quotaoff-start log items. */ -static struct xfs_item_ops xfs_qm_qoff_logitem_ops = { +static const struct xfs_item_ops xfs_qm_qoff_logitem_ops = { .iop_size = xfs_qm_qoff_logitem_size, .iop_format = xfs_qm_qoff_logitem_format, .iop_pin = xfs_qm_qoff_logitem_pin, diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index d22e62623437..35c2aff38b20 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -217,7 +217,7 @@ xfs_efi_item_committing( /* * This is the ops vector shared by all efi log items. */ -static struct xfs_item_ops xfs_efi_item_ops = { +static const struct xfs_item_ops xfs_efi_item_ops = { .iop_size = xfs_efi_item_size, .iop_format = xfs_efi_item_format, .iop_pin = xfs_efi_item_pin, @@ -477,7 +477,7 @@ xfs_efd_item_committing( /* * This is the ops vector shared by all efd log items. */ -static struct xfs_item_ops xfs_efd_item_ops = { +static const struct xfs_item_ops xfs_efd_item_ops = { .iop_size = xfs_efd_item_size, .iop_format = xfs_efd_item_format, .iop_pin = xfs_efd_item_pin, diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index b7cf21ba240f..abaafdbb3e65 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -795,7 +795,7 @@ xfs_inode_item_committing( /* * This is the ops vector shared by all buf log items. */ -static struct xfs_item_ops xfs_inode_item_ops = { +static const struct xfs_item_ops xfs_inode_item_ops = { .iop_size = xfs_inode_item_size, .iop_format = xfs_inode_item_format, .iop_pin = xfs_inode_item_pin, diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 2758a6277c52..a14cd89fe465 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -626,7 +626,7 @@ xfs_log_item_init( struct xfs_mount *mp, struct xfs_log_item *item, int type, - struct xfs_item_ops *ops) + const struct xfs_item_ops *ops) { item->li_mountp = mp; item->li_ailp = mp->m_ail; diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 78c9039994af..3f7bf451c034 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -137,7 +137,7 @@ struct xfs_trans; void xfs_log_item_init(struct xfs_mount *mp, struct xfs_log_item *item, int type, - struct xfs_item_ops *ops); + const struct xfs_item_ops *ops); xfs_lsn_t xfs_log_done(struct xfs_mount *mp, struct xlog_ticket *ticket, diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 603f3eb52041..3ae713c0abd9 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -326,7 +326,7 @@ typedef struct xfs_log_item { struct xfs_log_item *); /* buffer item iodone */ /* callback func */ - struct xfs_item_ops *li_ops; /* function list */ + const struct xfs_item_ops *li_ops; /* function list */ /* delayed logging */ struct list_head li_cil; /* CIL pointers */ @@ -341,7 +341,7 @@ typedef struct xfs_log_item { { XFS_LI_IN_AIL, "IN_AIL" }, \ { XFS_LI_ABORTED, "ABORTED" } -typedef struct xfs_item_ops { +struct xfs_item_ops { uint (*iop_size)(xfs_log_item_t *); void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *); void (*iop_pin)(xfs_log_item_t *); @@ -352,7 +352,7 @@ typedef struct xfs_item_ops { void (*iop_push)(xfs_log_item_t *); bool (*iop_pushbuf)(xfs_log_item_t *); void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t); -} xfs_item_ops_t; +}; #define IOP_SIZE(ip) (*(ip)->li_ops->iop_size)(ip) #define IOP_FORMAT(ip,vp) (*(ip)->li_ops->iop_format)(ip, vp) diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 4ecf2a549060..ce9268a2f56b 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -112,7 +112,7 @@ xfs_readlink( char *link) { xfs_mount_t *mp = ip->i_mount; - int pathlen; + xfs_fsize_t pathlen; int error = 0; trace_xfs_readlink(ip); @@ -122,13 +122,19 @@ xfs_readlink( xfs_ilock(ip, XFS_ILOCK_SHARED); - ASSERT(S_ISLNK(ip->i_d.di_mode)); - ASSERT(ip->i_d.di_size <= MAXPATHLEN); - pathlen = ip->i_d.di_size; if (!pathlen) goto out; + if (pathlen < 0 || pathlen > MAXPATHLEN) { + xfs_alert(mp, "%s: inode (%llu) bad symlink length (%lld)", + __func__, (unsigned long long) ip->i_ino, + (long long) pathlen); + ASSERT(0); + return XFS_ERROR(EFSCORRUPTED); + } + + if (ip->i_df.if_flags & XFS_IFINLINE) { memcpy(link, ip->i_df.if_u1.if_data, pathlen); link[pathlen] = '\0'; |