diff options
Diffstat (limited to 'fs')
134 files changed, 1888 insertions, 1103 deletions
diff --git a/fs/afs/addr_list.c b/fs/afs/addr_list.c index a936f9ea5610..63bf096b721a 100644 --- a/fs/afs/addr_list.c +++ b/fs/afs/addr_list.c @@ -298,8 +298,8 @@ int afs_merge_fs_addr4(struct afs_net *net, struct afs_addr_list *alist, srx.transport.sin.sin_addr.s_addr = xdr; peer = rxrpc_kernel_lookup_peer(net->socket, &srx, GFP_KERNEL); - if (!peer) - return -ENOMEM; + if (IS_ERR(peer)) + return PTR_ERR(peer); for (i = 0; i < alist->nr_ipv4; i++) { if (peer == alist->addrs[i].peer) { @@ -342,8 +342,8 @@ int afs_merge_fs_addr6(struct afs_net *net, struct afs_addr_list *alist, memcpy(&srx.transport.sin6.sin6_addr, xdr, 16); peer = rxrpc_kernel_lookup_peer(net->socket, &srx, GFP_KERNEL); - if (!peer) - return -ENOMEM; + if (IS_ERR(peer)) + return PTR_ERR(peer); for (i = alist->nr_ipv4; i < alist->nr_addrs; i++) { if (peer == alist->addrs[i].peer) { diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 8e89cc5b2820..fb857faaf0d6 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -47,6 +47,7 @@ #include <linux/dax.h> #include <linux/uaccess.h> #include <uapi/linux/rseq.h> +#include <linux/rseq.h> #include <asm/param.h> #include <asm/page.h> @@ -286,7 +287,7 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec, } #ifdef CONFIG_RSEQ NEW_AUX_ENT(AT_RSEQ_FEATURE_SIZE, offsetof(struct rseq, end)); - NEW_AUX_ENT(AT_RSEQ_ALIGN, __alignof__(struct rseq)); + NEW_AUX_ENT(AT_RSEQ_ALIGN, rseq_alloc_align()); #endif #undef NEW_AUX_ENT /* AT_NULL is zero; clear the rest too */ diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index c284f48cfae4..2a886bece810 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -3340,7 +3340,6 @@ again: btrfs_abort_transaction(trans, ret); goto out_put; } - WARN_ON(ret); /* We've already setup this transaction, go ahead and exit */ if (block_group->cache_generation == trans->transid && diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index d97bbbd045e0..56ff8afe9a22 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -1657,7 +1657,7 @@ int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans, if (unlikely(ret)) { btrfs_err(trans->fs_info, "failed to add delayed dir index item, root: %llu, inode: %llu, index: %llu, error: %d", - index, btrfs_root_id(node->root), node->inode_id, ret); + btrfs_root_id(node->root), node->inode_id, index, ret); btrfs_delayed_item_release_metadata(dir->root, item); btrfs_release_delayed_item(item); } diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index f6fa15a1193f..407830d86d0d 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1994,7 +1994,7 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, int level = btrfs_super_log_root_level(disk_super); if (unlikely(fs_devices->rw_devices == 0)) { - btrfs_warn(fs_info, "log replay required on RO media"); + btrfs_err(fs_info, "log replay required on RO media"); return -EIO; } @@ -2008,9 +2008,9 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, check.owner_root = BTRFS_TREE_LOG_OBJECTID; log_tree_root->node = read_tree_block(fs_info, bytenr, &check); if (IS_ERR(log_tree_root->node)) { - btrfs_warn(fs_info, "failed to read log tree"); ret = PTR_ERR(log_tree_root->node); log_tree_root->node = NULL; + btrfs_err(fs_info, "failed to read log tree with error: %d", ret); btrfs_put_root(log_tree_root); return ret; } @@ -2023,9 +2023,9 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, /* returns with log_tree_root freed on success */ ret = btrfs_recover_log_trees(log_tree_root); btrfs_put_root(log_tree_root); - if (ret) { - btrfs_handle_fs_error(fs_info, ret, - "Failed to recover log tree"); + if (unlikely(ret)) { + ASSERT(BTRFS_FS_ERROR(fs_info) != 0); + btrfs_err(fs_info, "failed to recover log trees with error: %d", ret); return ret; } @@ -2972,7 +2972,6 @@ static int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info) task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid"); if (IS_ERR(task)) { /* fs_info->update_uuid_tree_gen remains 0 in all error case */ - btrfs_warn(fs_info, "failed to start uuid_rescan task"); up(&fs_info->uuid_tree_rescan_sem); return PTR_ERR(task); } @@ -3188,7 +3187,7 @@ int btrfs_check_features(struct btrfs_fs_info *fs_info, bool is_rw_mount) if (incompat & ~BTRFS_FEATURE_INCOMPAT_SUPP) { btrfs_err(fs_info, "cannot mount because of unknown incompat features (0x%llx)", - incompat); + incompat & ~BTRFS_FEATURE_INCOMPAT_SUPP); return -EINVAL; } @@ -3220,7 +3219,7 @@ int btrfs_check_features(struct btrfs_fs_info *fs_info, bool is_rw_mount) if (compat_ro_unsupp && is_rw_mount) { btrfs_err(fs_info, "cannot mount read-write because of unknown compat_ro features (0x%llx)", - compat_ro); + compat_ro_unsupp); return -EINVAL; } @@ -3233,7 +3232,7 @@ int btrfs_check_features(struct btrfs_fs_info *fs_info, bool is_rw_mount) !btrfs_test_opt(fs_info, NOLOGREPLAY)) { btrfs_err(fs_info, "cannot replay dirty log with unsupported compat_ro features (0x%llx), try rescue=nologreplay", - compat_ro); + compat_ro_unsupp); return -EINVAL; } @@ -3595,7 +3594,6 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device } } - btrfs_zoned_reserve_data_reloc_bg(fs_info); btrfs_free_zone_cache(fs_info); btrfs_check_active_zone_reservation(fs_info); @@ -3623,6 +3621,12 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device goto fail_cleaner; } + /* + * Starts a transaction, must be called after the transaction kthread + * is initialized. + */ + btrfs_zoned_reserve_data_reloc_bg(fs_info); + ret = btrfs_read_qgroup_config(fs_info); if (ret) goto fail_trans_kthread; @@ -3642,7 +3646,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device fs_info->fs_root = btrfs_get_fs_root(fs_info, BTRFS_FS_TREE_OBJECTID, true); if (IS_ERR(fs_info->fs_root)) { ret = PTR_ERR(fs_info->fs_root); - btrfs_warn(fs_info, "failed to read fs tree: %d", ret); + btrfs_err(fs_info, "failed to read fs tree: %d", ret); fs_info->fs_root = NULL; goto fail_qgroup; } @@ -3663,8 +3667,7 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device btrfs_info(fs_info, "checking UUID tree"); ret = btrfs_check_uuid_tree(fs_info); if (ret) { - btrfs_warn(fs_info, - "failed to check the UUID tree: %d", ret); + btrfs_err(fs_info, "failed to check the UUID tree: %d", ret); close_ctree(fs_info); return ret; } @@ -4399,9 +4402,17 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) */ btrfs_flush_workqueue(fs_info->delayed_workers); - ret = btrfs_commit_super(fs_info); - if (ret) - btrfs_err(fs_info, "commit super ret %d", ret); + /* + * If the filesystem is shutdown, then an attempt to commit the + * super block (or any write) will just fail. Since we freeze + * the filesystem before shutting it down, the filesystem is in + * a consistent state and we don't need to commit super blocks. + */ + if (!btrfs_is_shutdown(fs_info)) { + ret = btrfs_commit_super(fs_info); + if (ret) + btrfs_err(fs_info, "commit super block returned %d", ret); + } } kthread_stop(fs_info->transaction_kthread); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 03cf9f242c70..b0d9baf5b412 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2933,9 +2933,15 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans) while (!TRANS_ABORTED(trans) && cached_state) { struct extent_state *next_state; - if (btrfs_test_opt(fs_info, DISCARD_SYNC)) + if (btrfs_test_opt(fs_info, DISCARD_SYNC)) { ret = btrfs_discard_extent(fs_info, start, end + 1 - start, NULL, true); + if (ret) { + btrfs_warn(fs_info, + "discard failed for extent [%llu, %llu]: errno=%d %s", + start, end, ret, btrfs_decode_error(ret)); + } + } next_state = btrfs_next_extent_state(unpin, cached_state); btrfs_clear_extent_dirty(unpin, start, end, &cached_state); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 744a1fff6eef..5f97a3d2a8d7 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4507,6 +4507,7 @@ static int try_release_subpage_extent_buffer(struct folio *folio) */ if (!test_and_clear_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags)) { spin_unlock(&eb->refs_lock); + rcu_read_lock(); break; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 6efb543f1c24..a6da98435ef7 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1392,10 +1392,25 @@ static int cow_one_range(struct btrfs_inode *inode, struct folio *locked_folio, return ret; free_reserved: + /* + * If we have reserved an extent for the current range and failed to + * create the respective extent map or ordered extent, it means that + * when we reserved the extent we decremented the extent's size from + * the data space_info's bytes_may_use counter and + * incremented the space_info's bytes_reserved counter by the same + * amount. + * + * We must make sure extent_clear_unlock_delalloc() does not try + * to decrement again the data space_info's bytes_may_use counter, which + * will be handled by btrfs_free_reserved_extent(). + * + * Therefore we do not pass it the flag EXTENT_CLEAR_DATA_RESV, but only + * EXTENT_CLEAR_META_RESV. + */ extent_clear_unlock_delalloc(inode, file_offset, cur_end, locked_folio, cached, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | - EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, + EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV, PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); btrfs_qgroup_free_data(inode, NULL, file_offset, cur_len, NULL); @@ -4764,7 +4779,7 @@ int btrfs_delete_subvolume(struct btrfs_inode *dir, struct dentry *dentry) spin_unlock(&dest->root_item_lock); btrfs_warn(fs_info, "attempt to delete subvolume %llu with active swapfile", - btrfs_root_id(root)); + btrfs_root_id(dest)); ret = -EPERM; goto out_up_write; } @@ -6597,6 +6612,25 @@ int btrfs_create_new_inode(struct btrfs_trans_handle *trans, int ret; bool xa_reserved = false; + if (!args->orphan && !args->subvol) { + /* + * Before anything else, check if we can add the name to the + * parent directory. We want to avoid a dir item overflow in + * case we have an existing dir item due to existing name + * hash collisions. We do this check here before we call + * btrfs_add_link() down below so that we can avoid a + * transaction abort (which could be exploited by malicious + * users). + * + * For subvolumes we already do this in btrfs_mksubvol(). + */ + ret = btrfs_check_dir_item_collision(BTRFS_I(dir)->root, + btrfs_ino(BTRFS_I(dir)), + name); + if (ret < 0) + return ret; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ae2173235c4d..b805dd9227ef 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -672,6 +672,13 @@ static noinline int create_subvol(struct mnt_idmap *idmap, goto out; } + /* + * Subvolumes have orphans cleaned on first dentry lookup. A new + * subvolume cannot have any orphans, so we should set the bit before we + * add the subvolume dentry to the dentry cache, so that it is in the + * same state as a subvolume after first lookup. + */ + set_bit(BTRFS_ROOT_ORPHAN_CLEANUP, &new_root->state); d_instantiate_new(dentry, new_inode_args.inode); new_inode_args.inode = NULL; @@ -3852,6 +3859,25 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, goto out; } + received_uuid_changed = memcmp(root_item->received_uuid, sa->uuid, + BTRFS_UUID_SIZE); + + /* + * Before we attempt to add the new received uuid, check if we have room + * for it in case there's already an item. If the size of the existing + * item plus this root's ID (u64) exceeds the maximum item size, we can + * return here without the need to abort a transaction. If we don't do + * this check, the btrfs_uuid_tree_add() call below would fail with + * -EOVERFLOW and result in a transaction abort. Malicious users could + * exploit this to turn the fs into RO mode. + */ + if (received_uuid_changed && !btrfs_is_empty_uuid(sa->uuid)) { + ret = btrfs_uuid_tree_check_overflow(fs_info, sa->uuid, + BTRFS_UUID_KEY_RECEIVED_SUBVOL); + if (ret < 0) + goto out; + } + /* * 1 - root item * 2 - uuid items (received uuid + subvol uuid) @@ -3867,15 +3893,12 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, sa->rtime.sec = ct.tv_sec; sa->rtime.nsec = ct.tv_nsec; - received_uuid_changed = memcmp(root_item->received_uuid, sa->uuid, - BTRFS_UUID_SIZE); if (received_uuid_changed && !btrfs_is_empty_uuid(root_item->received_uuid)) { ret = btrfs_uuid_tree_remove(trans, root_item->received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, btrfs_root_id(root)); if (unlikely(ret && ret != -ENOENT)) { - btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); goto out; } @@ -3890,7 +3913,8 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, ret = btrfs_update_root(trans, fs_info->tree_root, &root->root_key, &root->root_item); - if (ret < 0) { + if (unlikely(ret < 0)) { + btrfs_abort_transaction(trans, ret); btrfs_end_transaction(trans); goto out; } @@ -4581,7 +4605,7 @@ static int btrfs_uring_read_extent(struct kiocb *iocb, struct iov_iter *iter, { struct btrfs_inode *inode = BTRFS_I(file_inode(iocb->ki_filp)); struct extent_io_tree *io_tree = &inode->io_tree; - struct page **pages; + struct page **pages = NULL; struct btrfs_uring_priv *priv = NULL; unsigned long nr_pages; int ret; @@ -4639,6 +4663,11 @@ out_fail: btrfs_unlock_extent(io_tree, start, lockend, &cached_state); btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); kfree(priv); + for (int i = 0; i < nr_pages; i++) { + if (pages[i]) + __free_page(pages[i]); + } + kfree(pages); return ret; } diff --git a/fs/btrfs/messages.h b/fs/btrfs/messages.h index 943e53980945..c8e92efce405 100644 --- a/fs/btrfs/messages.h +++ b/fs/btrfs/messages.h @@ -31,9 +31,6 @@ void _btrfs_printk(const struct btrfs_fs_info *fs_info, unsigned int level, cons #define btrfs_printk_in_rcu(fs_info, level, fmt, args...) \ btrfs_no_printk(fs_info, fmt, ##args) -#define btrfs_printk_in_rcu(fs_info, level, fmt, args...) \ - btrfs_no_printk(fs_info, fmt, ##args) - #define btrfs_printk_rl_in_rcu(fs_info, level, fmt, args...) \ btrfs_no_printk(fs_info, fmt, ##args) diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c index f189bf09ce6a..b7dfe877cf8d 100644 --- a/fs/btrfs/print-tree.c +++ b/fs/btrfs/print-tree.c @@ -38,6 +38,7 @@ static const struct root_name_map root_map[] = { { BTRFS_BLOCK_GROUP_TREE_OBJECTID, "BLOCK_GROUP_TREE" }, { BTRFS_DATA_RELOC_TREE_OBJECTID, "DATA_RELOC_TREE" }, { BTRFS_RAID_STRIPE_TREE_OBJECTID, "RAID_STRIPE_TREE" }, + { BTRFS_REMAP_TREE_OBJECTID, "REMAP_TREE" }, }; const char *btrfs_root_name(const struct btrfs_key *key, char *buf) @@ -415,6 +416,9 @@ static void key_type_string(const struct btrfs_key *key, char *buf, int buf_size [BTRFS_UUID_KEY_SUBVOL] = "UUID_KEY_SUBVOL", [BTRFS_UUID_KEY_RECEIVED_SUBVOL] = "UUID_KEY_RECEIVED_SUBVOL", [BTRFS_RAID_STRIPE_KEY] = "RAID_STRIPE", + [BTRFS_IDENTITY_REMAP_KEY] = "IDENTITY_REMAP", + [BTRFS_REMAP_KEY] = "REMAP", + [BTRFS_REMAP_BACKREF_KEY] = "REMAP_BACKREF", }; if (key->type == 0 && key->objectid == BTRFS_FREE_SPACE_OBJECTID) @@ -435,6 +439,7 @@ void btrfs_print_leaf(const struct extent_buffer *l) struct btrfs_extent_data_ref *dref; struct btrfs_shared_data_ref *sref; struct btrfs_dev_extent *dev_extent; + struct btrfs_remap_item *remap; struct btrfs_key key; if (!l) @@ -569,6 +574,11 @@ void btrfs_print_leaf(const struct extent_buffer *l) print_raid_stripe_key(l, btrfs_item_size(l, i), btrfs_item_ptr(l, i, struct btrfs_stripe_extent)); break; + case BTRFS_REMAP_KEY: + case BTRFS_REMAP_BACKREF_KEY: + remap = btrfs_item_ptr(l, i, struct btrfs_remap_item); + pr_info("\t\taddress %llu\n", btrfs_remap_address(l, remap)); + break; } } } diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 3cdd9755dc52..3b2a6517d0b5 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -370,7 +370,7 @@ static bool squota_check_parent_usage(struct btrfs_fs_info *fs_info, struct btrf nr_members++; } mismatch = (parent->excl != excl_sum || parent->rfer != rfer_sum || - parent->excl_cmpr != excl_cmpr_sum || parent->rfer_cmpr != excl_cmpr_sum); + parent->excl_cmpr != excl_cmpr_sum || parent->rfer_cmpr != rfer_cmpr_sum); WARN(mismatch, "parent squota qgroup %hu/%llu has mismatched usage from its %d members. " diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 95db7c48fbad..b2343aed7a5d 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4399,6 +4399,8 @@ static int move_existing_remaps(struct btrfs_fs_info *fs_info, leaf = path->nodes[0]; } + + btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); } remap = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_remap_item); @@ -4723,6 +4725,7 @@ int btrfs_last_identity_remap_gone(struct btrfs_chunk_map *chunk_map, ret = btrfs_remove_dev_extents(trans, chunk_map); if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); return ret; } @@ -4732,6 +4735,7 @@ int btrfs_last_identity_remap_gone(struct btrfs_chunk_map *chunk_map, if (unlikely(ret)) { mutex_unlock(&trans->fs_info->chunk_mutex); btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); return ret; } } @@ -4750,6 +4754,7 @@ int btrfs_last_identity_remap_gone(struct btrfs_chunk_map *chunk_map, ret = remove_chunk_stripes(trans, chunk_map, path); if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); return ret; } @@ -5982,6 +5987,9 @@ static int remove_range_from_remap_tree(struct btrfs_trans_handle *trans, struct btrfs_block_group *dest_bg; dest_bg = btrfs_lookup_block_group(fs_info, new_addr); + if (unlikely(!dest_bg)) + return -EUCLEAN; + adjust_block_group_remap_bytes(trans, dest_bg, -overlap_length); btrfs_put_block_group(dest_bg); ret = btrfs_add_to_free_space_tree(trans, diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 81022d912abb..bc94bbc00772 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -743,7 +743,7 @@ static void scrub_verify_one_metadata(struct scrub_stripe *stripe, int sector_nr btrfs_warn_rl(fs_info, "scrub: tree block %llu mirror %u has bad fsid, has %pU want %pU", logical, stripe->mirror_num, - header->fsid, fs_info->fs_devices->fsid); + header->fsid, fs_info->fs_devices->metadata_uuid); return; } if (memcmp(header->chunk_tree_uuid, fs_info->chunk_tree_uuid, diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c index 52a267a5dd80..87cbc051cb12 100644 --- a/fs/btrfs/space-info.c +++ b/fs/btrfs/space-info.c @@ -2194,8 +2194,11 @@ void btrfs_reclaim_sweep(const struct btrfs_fs_info *fs_info) if (!btrfs_should_periodic_reclaim(space_info)) continue; for (raid = 0; raid < BTRFS_NR_RAID_TYPES; raid++) { - if (do_reclaim_sweep(space_info, raid)) + if (do_reclaim_sweep(space_info, raid)) { + spin_lock(&space_info->lock); btrfs_set_periodic_reclaim_ready(space_info, false); + spin_unlock(&space_info->lock); + } } } } diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 7ef8c9b7dfc1..8dd77c431974 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -1905,6 +1905,22 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, ret = btrfs_uuid_tree_add(trans, new_root_item->received_uuid, BTRFS_UUID_KEY_RECEIVED_SUBVOL, objectid); + /* + * We are creating of lot of snapshots of the same root that was + * received (has a received UUID) and reached a leaf's limit for + * an item. We can safely ignore this and avoid a transaction + * abort. A deletion of this snapshot will still work since we + * ignore if an item with a BTRFS_UUID_KEY_RECEIVED_SUBVOL key + * is missing (see btrfs_delete_subvolume()). Send/receive will + * work too since it peeks the first root id from the existing + * item (it could peek any), and in case it's missing it + * falls back to search by BTRFS_UUID_KEY_SUBVOL keys. + * Creation of a snapshot does not require CAP_SYS_ADMIN, so + * we don't want users triggering transaction aborts, either + * intentionally or not. + */ + if (ret == -EOVERFLOW) + ret = 0; if (unlikely(ret && ret != -EEXIST)) { btrfs_abort_transaction(trans, ret); goto fail; diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 452394b34d01..516ef62c8f43 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1284,7 +1284,7 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key, } if (unlikely(btrfs_root_drop_level(&ri) >= BTRFS_MAX_LEVEL)) { generic_err(leaf, slot, - "invalid root level, have %u expect [0, %u]", + "invalid root drop_level, have %u expect [0, %u]", btrfs_root_drop_level(&ri), BTRFS_MAX_LEVEL - 1); return -EUCLEAN; } @@ -1740,7 +1740,7 @@ static int check_extent_data_ref(struct extent_buffer *leaf, objectid > BTRFS_LAST_FREE_OBJECTID)) { extent_err(leaf, slot, "invalid extent data backref objectid value %llu", - root); + objectid); return -EUCLEAN; } if (unlikely(!IS_ALIGNED(offset, leaf->fs_info->sectorsize))) { @@ -1921,7 +1921,7 @@ static int check_dev_extent_item(const struct extent_buffer *leaf, if (unlikely(prev_key->offset + prev_len > key->offset)) { generic_err(leaf, slot, "dev extent overlap, prev offset %llu len %llu current offset %llu", - prev_key->objectid, prev_len, key->offset); + prev_key->offset, prev_len, key->offset); return -EUCLEAN; } } diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 780a06d59240..552fef3c385a 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -6195,6 +6195,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_log_ctx *ctx) { + const bool orig_log_new_dentries = ctx->log_new_dentries; int ret = 0; /* @@ -6256,7 +6257,11 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, * dir index key range logged for the directory. So we * must make sure the deletion is recorded. */ + ctx->log_new_dentries = false; ret = btrfs_log_inode(trans, inode, LOG_INODE_ALL, ctx); + if (!ret && ctx->log_new_dentries) + ret = log_new_dir_dentries(trans, inode, ctx); + btrfs_add_delayed_iput(inode); if (ret) break; @@ -6291,6 +6296,7 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, break; } + ctx->log_new_dentries = orig_log_new_dentries; ctx->logging_conflict_inodes = false; if (ret) free_conflicting_inodes(ctx); diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c index f24c14b9bb2f..43c17a1d3451 100644 --- a/fs/btrfs/uuid-tree.c +++ b/fs/btrfs/uuid-tree.c @@ -199,6 +199,44 @@ int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, const u8 *uuid, u8 return 0; } +/* + * Check if we can add one root ID to a UUID key. + * If the key does not yet exists, we can, otherwise only if extended item does + * not exceeds the maximum item size permitted by the leaf size. + * + * Returns 0 on success, negative value on error. + */ +int btrfs_uuid_tree_check_overflow(struct btrfs_fs_info *fs_info, + const u8 *uuid, u8 type) +{ + BTRFS_PATH_AUTO_FREE(path); + int ret; + u32 item_size; + struct btrfs_key key; + + if (WARN_ON_ONCE(!fs_info->uuid_root)) + return -EINVAL; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + btrfs_uuid_to_key(uuid, type, &key); + ret = btrfs_search_slot(NULL, fs_info->uuid_root, &key, path, 0, 0); + if (ret < 0) + return ret; + if (ret > 0) + return 0; + + item_size = btrfs_item_size(path->nodes[0], path->slots[0]); + + if (sizeof(struct btrfs_item) + item_size + sizeof(u64) > + BTRFS_LEAF_DATA_SIZE(fs_info)) + return -EOVERFLOW; + + return 0; +} + static int btrfs_uuid_iter_rem(struct btrfs_root *uuid_root, u8 *uuid, u8 type, u64 subid) { diff --git a/fs/btrfs/uuid-tree.h b/fs/btrfs/uuid-tree.h index c60ad20325cc..02b235a3653f 100644 --- a/fs/btrfs/uuid-tree.h +++ b/fs/btrfs/uuid-tree.h @@ -12,6 +12,8 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, const u8 *uuid, u8 typ u64 subid); int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, const u8 *uuid, u8 type, u64 subid); +int btrfs_uuid_tree_check_overflow(struct btrfs_fs_info *fs_info, + const u8 *uuid, u8 type); int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info); int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info); int btrfs_uuid_scan_kthread(void *data); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 6fb0c4cd50ff..be8975ef8b24 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -3587,7 +3587,7 @@ int btrfs_relocate_chunk(struct btrfs_fs_info *fs_info, u64 chunk_offset, bool v /* step one, relocate all the extents inside this chunk */ btrfs_scrub_pause(fs_info); - ret = btrfs_relocate_block_group(fs_info, chunk_offset, true); + ret = btrfs_relocate_block_group(fs_info, chunk_offset, verbose); btrfs_scrub_continue(fs_info); if (ret) { /* @@ -6907,7 +6907,7 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, ret = btrfs_translate_remap(fs_info, &new_logical, length); if (ret) - return ret; + goto out; if (new_logical != logical) { btrfs_free_chunk_map(map); @@ -6921,8 +6921,10 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, } num_copies = btrfs_chunk_map_num_copies(map); - if (io_geom.mirror_num > num_copies) - return -EINVAL; + if (io_geom.mirror_num > num_copies) { + ret = -EINVAL; + goto out; + } map_offset = logical - map->start; io_geom.raid56_full_stripe_start = (u64)-1; diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 39930d99943c..817ca4fb9efa 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -337,7 +337,10 @@ int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info) if (!btrfs_fs_incompat(fs_info, ZONED)) return 0; - mutex_lock(&fs_devices->device_list_mutex); + /* + * No need to take the device_list mutex here, we're still in the mount + * path and devices cannot be added to or removed from the list yet. + */ list_for_each_entry(device, &fs_devices->devices, dev_list) { /* We can skip reading of zone info for missing devices */ if (!device->bdev) @@ -347,7 +350,6 @@ int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info) if (ret) break; } - mutex_unlock(&fs_devices->device_list_mutex); return ret; } diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index e87b3bb94ee8..2090fc78529c 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -1326,7 +1326,6 @@ void ceph_process_folio_batch(struct address_space *mapping, continue; } else if (rc == -E2BIG) { folio_unlock(folio); - ceph_wbc->fbatch.folios[i] = NULL; break; } diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index f3fe786b4143..7dc307790240 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -79,7 +79,7 @@ static int mdsc_show(struct seq_file *s, void *p) if (req->r_inode) { seq_printf(s, " #%llx", ceph_ino(req->r_inode)); } else if (req->r_dentry) { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, req->r_dentry, &path_info, 0); if (IS_ERR(path)) path = NULL; @@ -98,7 +98,7 @@ static int mdsc_show(struct seq_file *s, void *p) } if (req->r_old_dentry) { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, req->r_old_dentry, &path_info, 0); if (IS_ERR(path)) path = NULL; diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 86d7aa594ea9..bac9cfb6b982 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1339,6 +1339,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) struct ceph_client *cl = fsc->client; struct ceph_mds_client *mdsc = fsc->mdsc; struct inode *inode = d_inode(dentry); + struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_mds_request *req; bool try_async = ceph_test_mount_opt(fsc, ASYNC_DIROPS); struct dentry *dn; @@ -1363,7 +1364,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry) if (!dn) { try_async = false; } else { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, dn, &path_info, 0); if (IS_ERR(path)) { try_async = false; @@ -1424,7 +1425,19 @@ retry: * We have enough caps, so we assume that the unlink * will succeed. Fix up the target inode and dcache. */ - drop_nlink(inode); + + /* + * Protect the i_nlink update with i_ceph_lock + * to precent racing against ceph_fill_inode() + * handling our completion on a worker thread + * and don't decrement if i_nlink has already + * been updated to zero by this completion. + */ + spin_lock(&ci->i_ceph_lock); + if (inode->i_nlink > 0) + drop_nlink(inode); + spin_unlock(&ci->i_ceph_lock); + d_delete(dentry); } else { spin_lock(&fsc->async_unlink_conflict_lock); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 66bbf6d517a9..5e7c73a29aa3 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -397,7 +397,7 @@ int ceph_open(struct inode *inode, struct file *file) if (!dentry) { do_sync = true; } else { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, dentry, &path_info, 0); if (IS_ERR(path)) { do_sync = true; @@ -807,7 +807,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, if (!dn) { try_async = false; } else { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, dn, &path_info, 0); if (IS_ERR(path)) { try_async = false; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index d76f9a79dc0c..d99e12d1100b 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -2551,7 +2551,7 @@ int __ceph_setattr(struct mnt_idmap *idmap, struct inode *inode, if (!dentry) { do_sync = true; } else { - struct ceph_path_info path_info; + struct ceph_path_info path_info = {0}; path = ceph_mdsc_build_path(mdsc, dentry, &path_info, 0); if (IS_ERR(path)) { do_sync = true; diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 23b6d00643c9..b1746273f186 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2768,6 +2768,7 @@ retry: if (ret < 0) { dput(parent); dput(cur); + __putname(path); return ERR_PTR(ret); } @@ -2777,6 +2778,7 @@ retry: if (len < 0) { dput(parent); dput(cur); + __putname(path); return ERR_PTR(len); } } @@ -2813,6 +2815,7 @@ retry: * cannot ever succeed. Creating paths that long is * possible with Ceph, but Linux cannot use them. */ + __putname(path); return ERR_PTR(-ENAMETOOLONG); } diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c index 4f86169c23f1..4b3d21402e10 100644 --- a/fs/erofs/inode.c +++ b/fs/erofs/inode.c @@ -222,6 +222,7 @@ err_out: static int erofs_fill_inode(struct inode *inode) { + const struct address_space_operations *aops; int err; trace_erofs_fill_inode(inode); @@ -254,7 +255,11 @@ static int erofs_fill_inode(struct inode *inode) } mapping_set_large_folios(inode->i_mapping); - return erofs_inode_set_aops(inode, inode, false); + aops = erofs_get_aops(inode, false); + if (IS_ERR(aops)) + return PTR_ERR(aops); + inode->i_mapping->a_ops = aops; + return 0; } /* diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index d1634455e389..a4f0a42cf8c3 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -471,26 +471,24 @@ static inline void *erofs_vm_map_ram(struct page **pages, unsigned int count) return NULL; } -static inline int erofs_inode_set_aops(struct inode *inode, - struct inode *realinode, bool no_fscache) +static inline const struct address_space_operations * +erofs_get_aops(struct inode *realinode, bool no_fscache) { if (erofs_inode_is_data_compressed(EROFS_I(realinode)->datalayout)) { if (!IS_ENABLED(CONFIG_EROFS_FS_ZIP)) - return -EOPNOTSUPP; + return ERR_PTR(-EOPNOTSUPP); DO_ONCE_LITE_IF(realinode->i_blkbits != PAGE_SHIFT, erofs_info, realinode->i_sb, "EXPERIMENTAL EROFS subpage compressed block support in use. Use at your own risk!"); - inode->i_mapping->a_ops = &z_erofs_aops; - return 0; + return &z_erofs_aops; } - inode->i_mapping->a_ops = &erofs_aops; if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND) && !no_fscache && erofs_is_fscache_mode(realinode->i_sb)) - inode->i_mapping->a_ops = &erofs_fscache_access_aops; + return &erofs_fscache_access_aops; if (IS_ENABLED(CONFIG_EROFS_FS_BACKED_BY_FILE) && erofs_is_fileio_mode(EROFS_SB(realinode->i_sb))) - inode->i_mapping->a_ops = &erofs_fileio_aops; - return 0; + return &erofs_fileio_aops; + return &erofs_aops; } int erofs_register_sysfs(struct super_block *sb); diff --git a/fs/erofs/ishare.c b/fs/erofs/ishare.c index ce980320a8b9..829d50d5c717 100644 --- a/fs/erofs/ishare.c +++ b/fs/erofs/ishare.c @@ -40,10 +40,14 @@ bool erofs_ishare_fill_inode(struct inode *inode) { struct erofs_sb_info *sbi = EROFS_SB(inode->i_sb); struct erofs_inode *vi = EROFS_I(inode); + const struct address_space_operations *aops; struct erofs_inode_fingerprint fp; struct inode *sharedinode; unsigned long hash; + aops = erofs_get_aops(inode, true); + if (IS_ERR(aops)) + return false; if (erofs_xattr_fill_inode_fingerprint(&fp, inode, sbi->domain_id)) return false; hash = xxh32(fp.opaque, fp.size, 0); @@ -56,15 +60,15 @@ bool erofs_ishare_fill_inode(struct inode *inode) } if (inode_state_read_once(sharedinode) & I_NEW) { - if (erofs_inode_set_aops(sharedinode, inode, true)) { - iget_failed(sharedinode); - kfree(fp.opaque); - return false; - } + sharedinode->i_mapping->a_ops = aops; sharedinode->i_size = vi->vfs_inode.i_size; unlock_new_inode(sharedinode); } else { kfree(fp.opaque); + if (aops != sharedinode->i_mapping->a_ops) { + iput(sharedinode); + return false; + } if (sharedinode->i_size != vi->vfs_inode.i_size) { _erofs_printk(inode->i_sb, KERN_WARNING "size(%lld:%lld) not matches for the same fingerprint\n", diff --git a/fs/erofs/super.c b/fs/erofs/super.c index d4995686ac6c..972a0c82198d 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -424,26 +424,23 @@ static const struct fs_parameter_spec erofs_fs_parameters[] = { static bool erofs_fc_set_dax_mode(struct fs_context *fc, unsigned int mode) { -#ifdef CONFIG_FS_DAX - struct erofs_sb_info *sbi = fc->s_fs_info; - - switch (mode) { - case EROFS_MOUNT_DAX_ALWAYS: - set_opt(&sbi->opt, DAX_ALWAYS); - clear_opt(&sbi->opt, DAX_NEVER); - return true; - case EROFS_MOUNT_DAX_NEVER: - set_opt(&sbi->opt, DAX_NEVER); - clear_opt(&sbi->opt, DAX_ALWAYS); - return true; - default: + if (IS_ENABLED(CONFIG_FS_DAX)) { + struct erofs_sb_info *sbi = fc->s_fs_info; + + if (mode == EROFS_MOUNT_DAX_ALWAYS) { + set_opt(&sbi->opt, DAX_ALWAYS); + clear_opt(&sbi->opt, DAX_NEVER); + return true; + } else if (mode == EROFS_MOUNT_DAX_NEVER) { + set_opt(&sbi->opt, DAX_NEVER); + clear_opt(&sbi->opt, DAX_ALWAYS); + return true; + } DBG_BUGON(1); return false; } -#else errorfc(fc, "dax options not supported"); return false; -#endif } static int erofs_fc_parse_param(struct fs_context *fc, @@ -460,31 +457,26 @@ static int erofs_fc_parse_param(struct fs_context *fc, switch (opt) { case Opt_user_xattr: -#ifdef CONFIG_EROFS_FS_XATTR - if (result.boolean) + if (!IS_ENABLED(CONFIG_EROFS_FS_XATTR)) + errorfc(fc, "{,no}user_xattr options not supported"); + else if (result.boolean) set_opt(&sbi->opt, XATTR_USER); else clear_opt(&sbi->opt, XATTR_USER); -#else - errorfc(fc, "{,no}user_xattr options not supported"); -#endif break; case Opt_acl: -#ifdef CONFIG_EROFS_FS_POSIX_ACL - if (result.boolean) + if (!IS_ENABLED(CONFIG_EROFS_FS_POSIX_ACL)) + errorfc(fc, "{,no}acl options not supported"); + else if (result.boolean) set_opt(&sbi->opt, POSIX_ACL); else clear_opt(&sbi->opt, POSIX_ACL); -#else - errorfc(fc, "{,no}acl options not supported"); -#endif break; case Opt_cache_strategy: -#ifdef CONFIG_EROFS_FS_ZIP - sbi->opt.cache_strategy = result.uint_32; -#else - errorfc(fc, "compression not supported, cache_strategy ignored"); -#endif + if (!IS_ENABLED(CONFIG_EROFS_FS_ZIP)) + errorfc(fc, "compression not supported, cache_strategy ignored"); + else + sbi->opt.cache_strategy = result.uint_32; break; case Opt_dax: if (!erofs_fc_set_dax_mode(fc, EROFS_MOUNT_DAX_ALWAYS)) @@ -533,24 +525,21 @@ static int erofs_fc_parse_param(struct fs_context *fc, break; #endif case Opt_directio: -#ifdef CONFIG_EROFS_FS_BACKED_BY_FILE - if (result.boolean) + if (!IS_ENABLED(CONFIG_EROFS_FS_BACKED_BY_FILE)) + errorfc(fc, "%s option not supported", erofs_fs_parameters[opt].name); + else if (result.boolean) set_opt(&sbi->opt, DIRECT_IO); else clear_opt(&sbi->opt, DIRECT_IO); -#else - errorfc(fc, "%s option not supported", erofs_fs_parameters[opt].name); -#endif break; case Opt_fsoffset: sbi->dif0.fsoff = result.uint_64; break; case Opt_inode_share: -#ifdef CONFIG_EROFS_FS_PAGE_CACHE_SHARE - set_opt(&sbi->opt, INODE_SHARE); -#else - errorfc(fc, "%s option not supported", erofs_fs_parameters[opt].name); -#endif + if (!IS_ENABLED(CONFIG_EROFS_FS_PAGE_CACHE_SHARE)) + errorfc(fc, "%s option not supported", erofs_fs_parameters[opt].name); + else + set_opt(&sbi->opt, INODE_SHARE); break; } return 0; @@ -809,8 +798,7 @@ static int erofs_fc_get_tree(struct fs_context *fc) ret = get_tree_bdev_flags(fc, erofs_fc_fill_super, IS_ENABLED(CONFIG_EROFS_FS_BACKED_BY_FILE) ? GET_TREE_BDEV_QUIET_LOOKUP : 0); -#ifdef CONFIG_EROFS_FS_BACKED_BY_FILE - if (ret == -ENOTBLK) { + if (IS_ENABLED(CONFIG_EROFS_FS_BACKED_BY_FILE) && ret == -ENOTBLK) { struct file *file; if (!fc->source) @@ -824,7 +812,6 @@ static int erofs_fc_get_tree(struct fs_context *fc) sbi->dif0.file->f_mapping->a_ops->read_folio) return get_tree_nodev(fc, erofs_fc_fill_super); } -#endif return ret; } @@ -1108,12 +1095,12 @@ static int erofs_show_options(struct seq_file *seq, struct dentry *root) seq_puts(seq, ",dax=never"); if (erofs_is_fileio_mode(sbi) && test_opt(opt, DIRECT_IO)) seq_puts(seq, ",directio"); -#ifdef CONFIG_EROFS_FS_ONDEMAND - if (sbi->fsid) - seq_printf(seq, ",fsid=%s", sbi->fsid); - if (sbi->domain_id) - seq_printf(seq, ",domain_id=%s", sbi->domain_id); -#endif + if (IS_ENABLED(CONFIG_EROFS_FS_ONDEMAND)) { + if (sbi->fsid) + seq_printf(seq, ",fsid=%s", sbi->fsid); + if (sbi->domain_id) + seq_printf(seq, ",domain_id=%s", sbi->domain_id); + } if (sbi->dif0.fsoff) seq_printf(seq, ",fsoffset=%llu", sbi->dif0.fsoff); if (test_opt(opt, INODE_SHARE)) diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index c8d8e129eb4b..30775502b56d 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -513,6 +513,7 @@ static int z_erofs_map_blocks_ext(struct inode *inode, unsigned int recsz = z_erofs_extent_recsize(vi->z_advise); erofs_off_t pos = round_up(Z_EROFS_MAP_HEADER_END(erofs_iloc(inode) + vi->inode_isize + vi->xattr_isize), recsz); + unsigned int bmask = sb->s_blocksize - 1; bool in_mbox = erofs_inode_in_metabox(inode); erofs_off_t lend = inode->i_size; erofs_off_t l, r, mid, pa, la, lstart; @@ -596,17 +597,17 @@ static int z_erofs_map_blocks_ext(struct inode *inode, map->m_flags |= EROFS_MAP_MAPPED | EROFS_MAP_FULL_MAPPED | EROFS_MAP_ENCODED; fmt = map->m_plen >> Z_EROFS_EXTENT_PLEN_FMT_BIT; + if (map->m_plen & Z_EROFS_EXTENT_PLEN_PARTIAL) + map->m_flags |= EROFS_MAP_PARTIAL_REF; + map->m_plen &= Z_EROFS_EXTENT_PLEN_MASK; if (fmt) map->m_algorithmformat = fmt - 1; - else if (interlaced && !erofs_blkoff(sb, map->m_pa)) + else if (interlaced && !((map->m_pa | map->m_plen) & bmask)) map->m_algorithmformat = Z_EROFS_COMPRESSION_INTERLACED; else map->m_algorithmformat = Z_EROFS_COMPRESSION_SHIFTED; - if (map->m_plen & Z_EROFS_EXTENT_PLEN_PARTIAL) - map->m_flags |= EROFS_MAP_PARTIAL_REF; - map->m_plen &= Z_EROFS_EXTENT_PLEN_MASK; } } map->m_llen = lend - map->m_la; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index a8c278c50083..5714e900567c 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -2061,7 +2061,8 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events, * @ep: the &struct eventpoll to be currently checked. * @depth: Current depth of the path being checked. * - * Return: depth of the subtree, or INT_MAX if we found a loop or went too deep. + * Return: depth of the subtree, or a value bigger than EP_MAX_NESTS if we found + * a loop or went too deep. */ static int ep_loop_check_proc(struct eventpoll *ep, int depth) { @@ -2080,7 +2081,7 @@ static int ep_loop_check_proc(struct eventpoll *ep, int depth) struct eventpoll *ep_tovisit; ep_tovisit = epi->ffd.file->private_data; if (ep_tovisit == inserting_into || depth > EP_MAX_NESTS) - result = INT_MAX; + result = EP_MAX_NESTS+1; else result = max(result, ep_loop_check_proc(ep_tovisit, depth + 1) + 1); if (result > EP_MAX_NESTS) diff --git a/fs/file_attr.c b/fs/file_attr.c index 6d2a298a786d..da983e105d70 100644 --- a/fs/file_attr.c +++ b/fs/file_attr.c @@ -378,7 +378,7 @@ SYSCALL_DEFINE5(file_getattr, int, dfd, const char __user *, filename, struct path filepath __free(path_put) = {}; unsigned int lookup_flags = 0; struct file_attr fattr; - struct file_kattr fa; + struct file_kattr fa = { .flags_valid = true }; /* hint only */ int error; BUILD_BUG_ON(sizeof(struct file_attr) < FILE_ATTR_SIZE_VER0); diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 8f8069fb76ba..7c75ed7e8979 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -198,10 +198,11 @@ static void wb_queue_work(struct bdi_writeback *wb, static bool wb_wait_for_completion_cb(struct wb_completion *done) { + unsigned long timeout = sysctl_hung_task_timeout_secs; unsigned long waited_secs = (jiffies - done->wait_start) / HZ; done->progress_stamp = jiffies; - if (waited_secs > sysctl_hung_task_timeout_secs) + if (timeout && (waited_secs > timeout)) pr_info("INFO: The task %s:%d has been waiting for writeback " "completion for more than %lu seconds.", current->comm, current->pid, waited_secs); @@ -1954,6 +1955,7 @@ static long writeback_sb_inodes(struct super_block *sb, .range_end = LLONG_MAX, }; unsigned long start_time = jiffies; + unsigned long timeout = sysctl_hung_task_timeout_secs; long write_chunk; long total_wrote = 0; /* count both pages and inodes */ unsigned long dirtied_before = jiffies; @@ -2040,9 +2042,8 @@ static long writeback_sb_inodes(struct super_block *sb, __writeback_single_inode(inode, &wbc); /* Report progress to inform the hung task detector of the progress. */ - if (work->done && work->done->progress_stamp && - (jiffies - work->done->progress_stamp) > HZ * - sysctl_hung_task_timeout_secs / 2) + if (work->done && work->done->progress_stamp && timeout && + (jiffies - work->done->progress_stamp) > HZ * timeout / 2) wake_up_all(work->done->waitq); wbc_detach_inode(&wbc); diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index a0c46aadb97d..00f0efaf12b2 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -80,18 +80,27 @@ static void iomap_set_range_uptodate(struct folio *folio, size_t off, { struct iomap_folio_state *ifs = folio->private; unsigned long flags; - bool uptodate = true; + bool mark_uptodate = true; if (folio_test_uptodate(folio)) return; if (ifs) { spin_lock_irqsave(&ifs->state_lock, flags); - uptodate = ifs_set_range_uptodate(folio, ifs, off, len); + /* + * If a read with bytes pending is in progress, we must not call + * folio_mark_uptodate(). The read completion path + * (iomap_read_end()) will call folio_end_read(), which uses XOR + * semantics to set the uptodate bit. If we set it here, the XOR + * in folio_end_read() will clear it, leaving the folio not + * uptodate. + */ + mark_uptodate = ifs_set_range_uptodate(folio, ifs, off, len) && + !ifs->read_bytes_pending; spin_unlock_irqrestore(&ifs->state_lock, flags); } - if (uptodate) + if (mark_uptodate) folio_mark_uptodate(folio); } @@ -624,6 +633,7 @@ static int iomap_readahead_iter(struct iomap_iter *iter, * iomap_readahead - Attempt to read pages from a file. * @ops: The operations vector for the filesystem. * @ctx: The ctx used for issuing readahead. + * @private: The filesystem-specific information for issuing iomap_iter. * * This function is for filesystems to call to implement their readahead * address_space operation. diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index 95254aa1b654..e911daedff65 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -87,6 +87,19 @@ static inline enum fserror_type iomap_dio_err_type(const struct iomap_dio *dio) return FSERR_DIRECTIO_READ; } +static inline bool should_report_dio_fserror(const struct iomap_dio *dio) +{ + switch (dio->error) { + case 0: + case -EAGAIN: + case -ENOTBLK: + /* don't send fsnotify for success or magic retry codes */ + return false; + default: + return true; + } +} + ssize_t iomap_dio_complete(struct iomap_dio *dio) { const struct iomap_dio_ops *dops = dio->dops; @@ -96,7 +109,7 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio) if (dops && dops->end_io) ret = dops->end_io(iocb, dio->size, ret, dio->flags); - if (dio->error) + if (should_report_dio_fserror(dio)) fserror_report_io(file_inode(iocb->ki_filp), iomap_dio_err_type(dio), offset, dio->size, dio->error, GFP_NOFS); diff --git a/fs/iomap/ioend.c b/fs/iomap/ioend.c index e4d57cb969f1..60546fa14dfe 100644 --- a/fs/iomap/ioend.c +++ b/fs/iomap/ioend.c @@ -69,11 +69,57 @@ static u32 iomap_finish_ioend_buffered(struct iomap_ioend *ioend) return folio_count; } +static DEFINE_SPINLOCK(failed_ioend_lock); +static LIST_HEAD(failed_ioend_list); + +static void +iomap_fail_ioends( + struct work_struct *work) +{ + struct iomap_ioend *ioend; + struct list_head tmp; + unsigned long flags; + + spin_lock_irqsave(&failed_ioend_lock, flags); + list_replace_init(&failed_ioend_list, &tmp); + spin_unlock_irqrestore(&failed_ioend_lock, flags); + + while ((ioend = list_first_entry_or_null(&tmp, struct iomap_ioend, + io_list))) { + list_del_init(&ioend->io_list); + iomap_finish_ioend_buffered(ioend); + cond_resched(); + } +} + +static DECLARE_WORK(failed_ioend_work, iomap_fail_ioends); + +static void iomap_fail_ioend_buffered(struct iomap_ioend *ioend) +{ + unsigned long flags; + + /* + * Bounce I/O errors to a workqueue to avoid nested i_lock acquisitions + * in the fserror code. The caller no longer owns the ioend reference + * after the spinlock drops. + */ + spin_lock_irqsave(&failed_ioend_lock, flags); + if (list_empty(&failed_ioend_list)) + WARN_ON_ONCE(!schedule_work(&failed_ioend_work)); + list_add_tail(&ioend->io_list, &failed_ioend_list); + spin_unlock_irqrestore(&failed_ioend_lock, flags); +} + static void ioend_writeback_end_bio(struct bio *bio) { struct iomap_ioend *ioend = iomap_ioend_from_bio(bio); ioend->io_error = blk_status_to_errno(bio->bi_status); + if (ioend->io_error) { + iomap_fail_ioend_buffered(ioend); + return; + } + iomap_finish_ioend_buffered(ioend); } @@ -169,17 +215,18 @@ ssize_t iomap_add_to_ioend(struct iomap_writepage_ctx *wpc, struct folio *folio, WARN_ON_ONCE(!folio->private && map_len < dirty_len); switch (wpc->iomap.type) { - case IOMAP_INLINE: - WARN_ON_ONCE(1); - return -EIO; + case IOMAP_UNWRITTEN: + ioend_flags |= IOMAP_IOEND_UNWRITTEN; + break; + case IOMAP_MAPPED: + break; case IOMAP_HOLE: return map_len; default: - break; + WARN_ON_ONCE(1); + return -EIO; } - if (wpc->iomap.type == IOMAP_UNWRITTEN) - ioend_flags |= IOMAP_IOEND_UNWRITTEN; if (wpc->iomap.flags & IOMAP_F_SHARED) ioend_flags |= IOMAP_IOEND_SHARED; if (folio_test_dropbehind(folio)) diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c index 7da66ca184f4..abec438330a7 100644 --- a/fs/minix/bitmap.c +++ b/fs/minix/bitmap.c @@ -247,7 +247,7 @@ struct inode *minix_new_inode(const struct inode *dir, umode_t mode) j += i * bits_per_zone; if (!j || j > sbi->s_ninodes) { iput(inode); - return ERR_PTR(-ENOSPC); + return ERR_PTR(-EFSCORRUPTED); } inode_init_owner(&nop_mnt_idmap, inode, dir, mode); inode->i_ino = j; diff --git a/fs/namespace.c b/fs/namespace.c index ebe19ded293a..854f4fc66469 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1531,23 +1531,33 @@ static struct mount *mnt_find_id_at_reverse(struct mnt_namespace *ns, u64 mnt_id static void *m_start(struct seq_file *m, loff_t *pos) { struct proc_mounts *p = m->private; + struct mount *mnt; down_read(&namespace_sem); - return mnt_find_id_at(p->ns, *pos); + mnt = mnt_find_id_at(p->ns, *pos); + if (mnt) + *pos = mnt->mnt_id_unique; + return mnt; } static void *m_next(struct seq_file *m, void *v, loff_t *pos) { - struct mount *next = NULL, *mnt = v; + struct mount *mnt = v; struct rb_node *node = rb_next(&mnt->mnt_node); - ++*pos; if (node) { - next = node_to_mount(node); + struct mount *next = node_to_mount(node); *pos = next->mnt_id_unique; + return next; } - return next; + + /* + * No more mounts. Set pos past current mount's ID so that if + * iteration restarts, mnt_find_id_at() returns NULL. + */ + *pos = mnt->mnt_id_unique + 1; + return NULL; } static void m_stop(struct seq_file *m, void *v) @@ -2791,7 +2801,8 @@ static inline void unlock_mount(struct pinned_mountpoint *m) } static void lock_mount_exact(const struct path *path, - struct pinned_mountpoint *mp); + struct pinned_mountpoint *mp, bool copy_mount, + unsigned int copy_flags); #define LOCK_MOUNT_MAYBE_BENEATH(mp, path, beneath) \ struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \ @@ -2799,7 +2810,10 @@ static void lock_mount_exact(const struct path *path, #define LOCK_MOUNT(mp, path) LOCK_MOUNT_MAYBE_BENEATH(mp, (path), false) #define LOCK_MOUNT_EXACT(mp, path) \ struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \ - lock_mount_exact((path), &mp) + lock_mount_exact((path), &mp, false, 0) +#define LOCK_MOUNT_EXACT_COPY(mp, path, copy_flags) \ + struct pinned_mountpoint mp __cleanup(unlock_mount) = {}; \ + lock_mount_exact((path), &mp, true, (copy_flags)) static int graft_tree(struct mount *mnt, const struct pinned_mountpoint *mp) { @@ -3073,16 +3087,13 @@ static struct file *open_detached_copy(struct path *path, unsigned int flags) return file; } -DEFINE_FREE(put_empty_mnt_ns, struct mnt_namespace *, - if (!IS_ERR_OR_NULL(_T)) free_mnt_ns(_T)) - static struct mnt_namespace *create_new_namespace(struct path *path, unsigned int flags) { - struct mnt_namespace *new_ns __free(put_empty_mnt_ns) = NULL; - struct path to_path __free(path_put) = {}; struct mnt_namespace *ns = current->nsproxy->mnt_ns; struct user_namespace *user_ns = current_user_ns(); - struct mount *new_ns_root; + struct mnt_namespace *new_ns; + struct mount *new_ns_root, *old_ns_root; + struct path to_path; struct mount *mnt; unsigned int copy_flags = 0; bool locked = false; @@ -3094,71 +3105,63 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned in if (IS_ERR(new_ns)) return ERR_CAST(new_ns); - scoped_guard(namespace_excl) { - new_ns_root = clone_mnt(ns->root, ns->root->mnt.mnt_root, copy_flags); - if (IS_ERR(new_ns_root)) - return ERR_CAST(new_ns_root); + old_ns_root = ns->root; + to_path.mnt = &old_ns_root->mnt; + to_path.dentry = old_ns_root->mnt.mnt_root; - /* - * If the real rootfs had a locked mount on top of it somewhere - * in the stack, lock the new mount tree as well so it can't be - * exposed. - */ - mnt = ns->root; - while (mnt->overmount) { - mnt = mnt->overmount; - if (mnt->mnt.mnt_flags & MNT_LOCKED) - locked = true; - } + VFS_WARN_ON_ONCE(old_ns_root->mnt.mnt_sb->s_type != &nullfs_fs_type); + + LOCK_MOUNT_EXACT_COPY(mp, &to_path, copy_flags); + if (IS_ERR(mp.parent)) { + free_mnt_ns(new_ns); + return ERR_CAST(mp.parent); } + new_ns_root = mp.parent; /* - * We dropped the namespace semaphore so we can actually lock - * the copy for mounting. The copied mount isn't attached to any - * mount namespace and it is thus excluded from any propagation. - * So realistically we're isolated and the mount can't be - * overmounted. + * If the real rootfs had a locked mount on top of it somewhere + * in the stack, lock the new mount tree as well so it can't be + * exposed. */ - - /* Borrow the reference from clone_mnt(). */ - to_path.mnt = &new_ns_root->mnt; - to_path.dentry = dget(new_ns_root->mnt.mnt_root); - - /* Now lock for actual mounting. */ - LOCK_MOUNT_EXACT(mp, &to_path); - if (unlikely(IS_ERR(mp.parent))) - return ERR_CAST(mp.parent); + mnt = old_ns_root; + while (mnt->overmount) { + mnt = mnt->overmount; + if (mnt->mnt.mnt_flags & MNT_LOCKED) + locked = true; + } /* - * We don't emulate unshare()ing a mount namespace. We stick to the - * restrictions of creating detached bind-mounts. It has a lot - * saner and simpler semantics. + * We don't emulate unshare()ing a mount namespace. We stick + * to the restrictions of creating detached bind-mounts. It + * has a lot saner and simpler semantics. */ mnt = __do_loopback(path, flags, copy_flags); - if (IS_ERR(mnt)) - return ERR_CAST(mnt); - scoped_guard(mount_writer) { + if (IS_ERR(mnt)) { + emptied_ns = new_ns; + umount_tree(new_ns_root, 0); + return ERR_CAST(mnt); + } + if (locked) mnt->mnt.mnt_flags |= MNT_LOCKED; /* - * Now mount the detached tree on top of the copy of the - * real rootfs we created. + * now mount the detached tree on top of the copy + * of the real rootfs we created. */ attach_mnt(mnt, new_ns_root, mp.mp); if (user_ns != ns->user_ns) lock_mnt_tree(new_ns_root); } - /* Add all mounts to the new namespace. */ - for (struct mount *p = new_ns_root; p; p = next_mnt(p, new_ns_root)) { - mnt_add_to_ns(new_ns, p); + for (mnt = new_ns_root; mnt; mnt = next_mnt(mnt, new_ns_root)) { + mnt_add_to_ns(new_ns, mnt); new_ns->nr_mounts++; } - new_ns->root = real_mount(no_free_ptr(to_path.mnt)); + new_ns->root = new_ns_root; ns_tree_add_raw(new_ns); - return no_free_ptr(new_ns); + return new_ns; } static struct file *open_new_namespace(struct path *path, unsigned int flags) @@ -3840,16 +3843,20 @@ static int do_new_mount(const struct path *path, const char *fstype, } static void lock_mount_exact(const struct path *path, - struct pinned_mountpoint *mp) + struct pinned_mountpoint *mp, bool copy_mount, + unsigned int copy_flags) { struct dentry *dentry = path->dentry; int err; + /* Assert that inode_lock() locked the correct inode. */ + VFS_WARN_ON_ONCE(copy_mount && !path_mounted(path)); + inode_lock(dentry->d_inode); namespace_lock(); if (unlikely(cant_mount(dentry))) err = -ENOENT; - else if (path_overmounted(path)) + else if (!copy_mount && path_overmounted(path)) err = -EBUSY; else err = get_mountpoint(dentry, mp); @@ -3857,9 +3864,15 @@ static void lock_mount_exact(const struct path *path, namespace_unlock(); inode_unlock(dentry->d_inode); mp->parent = ERR_PTR(err); - } else { - mp->parent = real_mount(path->mnt); + return; } + + if (copy_mount) + mp->parent = clone_mnt(real_mount(path->mnt), dentry, copy_flags); + else + mp->parent = real_mount(path->mnt); + if (unlikely(IS_ERR(mp->parent))) + __unlock_mount(mp); } int finish_automount(struct vfsmount *__m, const struct path *path) @@ -5678,6 +5691,8 @@ static int do_statmount(struct kstatmount *s, u64 mnt_id, u64 mnt_ns_id, s->mnt = mnt_file->f_path.mnt; ns = real_mount(s->mnt)->mnt_ns; + if (IS_ERR(ns)) + return PTR_ERR(ns); if (!ns) /* * We can't set mount point and mnt_ns_id since we don't have a diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index a9d1c3b2c084..dd1451bf7543 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -10,6 +10,202 @@ #include "internal.h" /* + * Perform the cleanup rituals after an unbuffered write is complete. + */ +static void netfs_unbuffered_write_done(struct netfs_io_request *wreq) +{ + struct netfs_inode *ictx = netfs_inode(wreq->inode); + + _enter("R=%x", wreq->debug_id); + + /* Okay, declare that all I/O is complete. */ + trace_netfs_rreq(wreq, netfs_rreq_trace_write_done); + + if (!wreq->error) + netfs_update_i_size(ictx, &ictx->inode, wreq->start, wreq->transferred); + + if (wreq->origin == NETFS_DIO_WRITE && + wreq->mapping->nrpages) { + /* mmap may have got underfoot and we may now have folios + * locally covering the region we just wrote. Attempt to + * discard the folios, but leave in place any modified locally. + * ->write_iter() is prevented from interfering by the DIO + * counter. + */ + pgoff_t first = wreq->start >> PAGE_SHIFT; + pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; + + invalidate_inode_pages2_range(wreq->mapping, first, last); + } + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_end(wreq->inode); + + _debug("finished"); + netfs_wake_rreq_flag(wreq, NETFS_RREQ_IN_PROGRESS, netfs_rreq_trace_wake_ip); + /* As we cleared NETFS_RREQ_IN_PROGRESS, we acquired its ref. */ + + if (wreq->iocb) { + size_t written = umin(wreq->transferred, wreq->len); + + wreq->iocb->ki_pos += written; + if (wreq->iocb->ki_complete) { + trace_netfs_rreq(wreq, netfs_rreq_trace_ki_complete); + wreq->iocb->ki_complete(wreq->iocb, wreq->error ?: written); + } + wreq->iocb = VFS_PTR_POISON; + } + + netfs_clear_subrequests(wreq); +} + +/* + * Collect the subrequest results of unbuffered write subrequests. + */ +static void netfs_unbuffered_write_collect(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq) +{ + trace_netfs_collect_sreq(wreq, subreq); + + spin_lock(&wreq->lock); + list_del_init(&subreq->rreq_link); + spin_unlock(&wreq->lock); + + wreq->transferred += subreq->transferred; + iov_iter_advance(&wreq->buffer.iter, subreq->transferred); + + stream->collected_to = subreq->start + subreq->transferred; + wreq->collected_to = stream->collected_to; + netfs_put_subrequest(subreq, netfs_sreq_trace_put_done); + + trace_netfs_collect_stream(wreq, stream); + trace_netfs_collect_state(wreq, wreq->collected_to, 0); +} + +/* + * Write data to the server without going through the pagecache and without + * writing it to the local cache. We dispatch the subrequests serially and + * wait for each to complete before dispatching the next, lest we leave a gap + * in the data written due to a failure such as ENOSPC. We could, however + * attempt to do preparation such as content encryption for the next subreq + * whilst the current is in progress. + */ +static int netfs_unbuffered_write(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *subreq = NULL; + struct netfs_io_stream *stream = &wreq->io_streams[0]; + int ret; + + _enter("%llx", wreq->len); + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_begin(wreq->inode); + + stream->collected_to = wreq->start; + + for (;;) { + bool retry = false; + + if (!subreq) { + netfs_prepare_write(wreq, stream, wreq->start + wreq->transferred); + subreq = stream->construct; + stream->construct = NULL; + stream->front = NULL; + } + + /* Check if (re-)preparation failed. */ + if (unlikely(test_bit(NETFS_SREQ_FAILED, &subreq->flags))) { + netfs_write_subrequest_terminated(subreq, subreq->error); + wreq->error = subreq->error; + break; + } + + iov_iter_truncate(&subreq->io_iter, wreq->len - wreq->transferred); + if (!iov_iter_count(&subreq->io_iter)) + break; + + subreq->len = netfs_limit_iter(&subreq->io_iter, 0, + stream->sreq_max_len, + stream->sreq_max_segs); + iov_iter_truncate(&subreq->io_iter, subreq->len); + stream->submit_extendable_to = subreq->len; + + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + stream->issue_write(subreq); + + /* Async, need to wait. */ + netfs_wait_for_in_progress_stream(wreq, stream); + + if (test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + retry = true; + } else if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) { + ret = subreq->error; + wreq->error = ret; + netfs_see_subrequest(subreq, netfs_sreq_trace_see_failed); + subreq = NULL; + break; + } + ret = 0; + + if (!retry) { + netfs_unbuffered_write_collect(wreq, stream, subreq); + subreq = NULL; + if (wreq->transferred >= wreq->len) + break; + if (!wreq->iocb && signal_pending(current)) { + ret = wreq->transferred ? -EINTR : -ERESTARTSYS; + trace_netfs_rreq(wreq, netfs_rreq_trace_intr); + break; + } + continue; + } + + /* We need to retry the last subrequest, so first reset the + * iterator, taking into account what, if anything, we managed + * to transfer. + */ + subreq->error = -EAGAIN; + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + if (subreq->transferred > 0) + iov_iter_advance(&wreq->buffer.iter, subreq->transferred); + + if (stream->source == NETFS_UPLOAD_TO_SERVER && + wreq->netfs_ops->retry_request) + wreq->netfs_ops->retry_request(wreq, stream); + + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + __clear_bit(NETFS_SREQ_BOUNDARY, &subreq->flags); + __clear_bit(NETFS_SREQ_FAILED, &subreq->flags); + subreq->io_iter = wreq->buffer.iter; + subreq->start = wreq->start + wreq->transferred; + subreq->len = wreq->len - wreq->transferred; + subreq->transferred = 0; + subreq->retry_count += 1; + stream->sreq_max_len = UINT_MAX; + stream->sreq_max_segs = INT_MAX; + + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + stream->prepare_write(subreq); + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + netfs_stat(&netfs_n_wh_retry_write_subreq); + } + + netfs_unbuffered_write_done(wreq); + _leave(" = %d", ret); + return ret; +} + +static void netfs_unbuffered_write_async(struct work_struct *work) +{ + struct netfs_io_request *wreq = container_of(work, struct netfs_io_request, work); + + netfs_unbuffered_write(wreq); + netfs_put_request(wreq, netfs_rreq_trace_put_complete); +} + +/* * Perform an unbuffered write where we may have to do an RMW operation on an * encrypted file. This can also be used for direct I/O writes. */ @@ -70,35 +266,35 @@ ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter * */ wreq->buffer.iter = *iter; } + + wreq->len = iov_iter_count(&wreq->buffer.iter); } __set_bit(NETFS_RREQ_USE_IO_ITER, &wreq->flags); - if (async) - __set_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &wreq->flags); /* Copy the data into the bounce buffer and encrypt it. */ // TODO /* Dispatch the write. */ __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - if (async) - wreq->iocb = iocb; - wreq->len = iov_iter_count(&wreq->buffer.iter); - ret = netfs_unbuffered_write(wreq, is_sync_kiocb(iocb), wreq->len); - if (ret < 0) { - _debug("begin = %zd", ret); - goto out; - } - if (!async) { - ret = netfs_wait_for_write(wreq); - if (ret > 0) - iocb->ki_pos += ret; - } else { + if (async) { + INIT_WORK(&wreq->work, netfs_unbuffered_write_async); + wreq->iocb = iocb; + queue_work(system_dfl_wq, &wreq->work); ret = -EIOCBQUEUED; + } else { + ret = netfs_unbuffered_write(wreq); + if (ret < 0) { + _debug("begin = %zd", ret); + } else { + iocb->ki_pos += wreq->transferred; + ret = wreq->transferred ?: wreq->error; + } + + netfs_put_request(wreq, netfs_rreq_trace_put_complete); } -out: netfs_put_request(wreq, netfs_rreq_trace_put_return); return ret; diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index 4319611f5354..d436e20d3418 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -198,6 +198,9 @@ struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, struct file *file, loff_t start, enum netfs_io_origin origin); +void netfs_prepare_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start); void netfs_reissue_write(struct netfs_io_stream *stream, struct netfs_io_subrequest *subreq, struct iov_iter *source); @@ -212,7 +215,6 @@ int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_c struct folio **writethrough_cache); ssize_t netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, struct folio *writethrough_cache); -int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len); /* * write_retry.c diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c index 61eab34ea67e..83eb3dc1adf8 100644 --- a/fs/netfs/write_collect.c +++ b/fs/netfs/write_collect.c @@ -399,27 +399,6 @@ bool netfs_write_collection(struct netfs_io_request *wreq) ictx->ops->invalidate_cache(wreq); } - if ((wreq->origin == NETFS_UNBUFFERED_WRITE || - wreq->origin == NETFS_DIO_WRITE) && - !wreq->error) - netfs_update_i_size(ictx, &ictx->inode, wreq->start, wreq->transferred); - - if (wreq->origin == NETFS_DIO_WRITE && - wreq->mapping->nrpages) { - /* mmap may have got underfoot and we may now have folios - * locally covering the region we just wrote. Attempt to - * discard the folios, but leave in place any modified locally. - * ->write_iter() is prevented from interfering by the DIO - * counter. - */ - pgoff_t first = wreq->start >> PAGE_SHIFT; - pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; - invalidate_inode_pages2_range(wreq->mapping, first, last); - } - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_end(wreq->inode); - _debug("finished"); netfs_wake_rreq_flag(wreq, NETFS_RREQ_IN_PROGRESS, netfs_rreq_trace_wake_ip); /* As we cleared NETFS_RREQ_IN_PROGRESS, we acquired its ref. */ diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c index 34894da5a23e..437268f65640 100644 --- a/fs/netfs/write_issue.c +++ b/fs/netfs/write_issue.c @@ -154,9 +154,9 @@ EXPORT_SYMBOL(netfs_prepare_write_failed); * Prepare a write subrequest. We need to allocate a new subrequest * if we don't have one. */ -static void netfs_prepare_write(struct netfs_io_request *wreq, - struct netfs_io_stream *stream, - loff_t start) +void netfs_prepare_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start) { struct netfs_io_subrequest *subreq; struct iov_iter *wreq_iter = &wreq->buffer.iter; @@ -699,41 +699,6 @@ ssize_t netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_c } /* - * Write data to the server without going through the pagecache and without - * writing it to the local cache. - */ -int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len) -{ - struct netfs_io_stream *upload = &wreq->io_streams[0]; - ssize_t part; - loff_t start = wreq->start; - int error = 0; - - _enter("%zx", len); - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_begin(wreq->inode); - - while (len) { - // TODO: Prepare content encryption - - _debug("unbuffered %zx", len); - part = netfs_advance_write(wreq, upload, start, len, false); - start += part; - len -= part; - rolling_buffer_advance(&wreq->buffer, part); - if (test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) - netfs_wait_for_paused_write(wreq); - if (test_bit(NETFS_RREQ_FAILED, &wreq->flags)) - break; - } - - netfs_end_issue_write(wreq); - _leave(" = %d", error); - return error; -} - -/* * Write some of a pending folio data back to the server and/or the cache. */ static int netfs_write_folio_single(struct netfs_io_request *wreq, diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index 12cb0ca738af..6bb30543eff0 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -87,7 +87,7 @@ config NFS_V4 space programs which can be found in the Linux nfs-utils package, available from http://linux-nfs.org/. - If unsure, say Y. + If unsure, say N. config NFS_SWAP bool "Provide swap over NFS support" @@ -100,6 +100,7 @@ config NFS_SWAP config NFS_V4_0 bool "NFS client support for NFSv4.0" depends on NFS_V4 + default y help This option enables support for minor version 0 of the NFSv4 protocol (RFC 3530) in the kernel's NFS client. diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 3e2de45c95fe..be2aebf62056 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -392,8 +392,13 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, if (status != 0) goto out_release_acls; - if (d_alias) + if (d_alias) { + if (d_is_dir(d_alias)) { + status = -EISDIR; + goto out_dput; + } dentry = d_alias; + } /* When we created the file with exclusive semantics, make * sure we set the attributes afterwards. */ diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 8fdbba7cad96..8e8a76a44ff0 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -36,19 +36,30 @@ * second map contains a reference to the entry in the first map. */ +static struct workqueue_struct *nfsd_export_wq; + #define EXPKEY_HASHBITS 8 #define EXPKEY_HASHMAX (1 << EXPKEY_HASHBITS) #define EXPKEY_HASHMASK (EXPKEY_HASHMAX -1) -static void expkey_put(struct kref *ref) +static void expkey_release(struct work_struct *work) { - struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + struct svc_expkey *key = container_of(to_rcu_work(work), + struct svc_expkey, ek_rwork); if (test_bit(CACHE_VALID, &key->h.flags) && !test_bit(CACHE_NEGATIVE, &key->h.flags)) path_put(&key->ek_path); auth_domain_put(key->ek_client); - kfree_rcu(key, ek_rcu); + kfree(key); +} + +static void expkey_put(struct kref *ref) +{ + struct svc_expkey *key = container_of(ref, struct svc_expkey, h.ref); + + INIT_RCU_WORK(&key->ek_rwork, expkey_release); + queue_rcu_work(nfsd_export_wq, &key->ek_rwork); } static int expkey_upcall(struct cache_detail *cd, struct cache_head *h) @@ -353,11 +364,13 @@ static void export_stats_destroy(struct export_stats *stats) EXP_STATS_COUNTERS_NUM); } -static void svc_export_release(struct rcu_head *rcu_head) +static void svc_export_release(struct work_struct *work) { - struct svc_export *exp = container_of(rcu_head, struct svc_export, - ex_rcu); + struct svc_export *exp = container_of(to_rcu_work(work), + struct svc_export, ex_rwork); + path_put(&exp->ex_path); + auth_domain_put(exp->ex_client); nfsd4_fslocs_free(&exp->ex_fslocs); export_stats_destroy(exp->ex_stats); kfree(exp->ex_stats); @@ -369,9 +382,8 @@ static void svc_export_put(struct kref *ref) { struct svc_export *exp = container_of(ref, struct svc_export, h.ref); - path_put(&exp->ex_path); - auth_domain_put(exp->ex_client); - call_rcu(&exp->ex_rcu, svc_export_release); + INIT_RCU_WORK(&exp->ex_rwork, svc_export_release); + queue_rcu_work(nfsd_export_wq, &exp->ex_rwork); } static int svc_export_upcall(struct cache_detail *cd, struct cache_head *h) @@ -1479,6 +1491,36 @@ const struct seq_operations nfs_exports_op = { .show = e_show, }; +/** + * nfsd_export_wq_init - allocate the export release workqueue + * + * Called once at module load. The workqueue runs deferred svc_export and + * svc_expkey release work scheduled by queue_rcu_work() in the cache put + * callbacks. + * + * Return values: + * %0: workqueue allocated + * %-ENOMEM: allocation failed + */ +int nfsd_export_wq_init(void) +{ + nfsd_export_wq = alloc_workqueue("nfsd_export", WQ_UNBOUND, 0); + if (!nfsd_export_wq) + return -ENOMEM; + return 0; +} + +/** + * nfsd_export_wq_shutdown - drain and free the export release workqueue + * + * Called once at module unload. Per-namespace teardown in + * nfsd_export_shutdown() has already drained all deferred work. + */ +void nfsd_export_wq_shutdown(void) +{ + destroy_workqueue(nfsd_export_wq); +} + /* * Initialize the exports module. */ @@ -1540,6 +1582,9 @@ nfsd_export_shutdown(struct net *net) cache_unregister_net(nn->svc_expkey_cache, net); cache_unregister_net(nn->svc_export_cache, net); + /* Drain deferred export and expkey release work. */ + rcu_barrier(); + flush_workqueue(nfsd_export_wq); cache_destroy_net(nn->svc_expkey_cache, net); cache_destroy_net(nn->svc_export_cache, net); svcauth_unix_purge(net); diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index d2b09cd76145..b05399374574 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -7,6 +7,7 @@ #include <linux/sunrpc/cache.h> #include <linux/percpu_counter.h> +#include <linux/workqueue.h> #include <uapi/linux/nfsd/export.h> #include <linux/nfs4.h> @@ -75,7 +76,7 @@ struct svc_export { u32 ex_layout_types; struct nfsd4_deviceid_map *ex_devid_map; struct cache_detail *cd; - struct rcu_head ex_rcu; + struct rcu_work ex_rwork; unsigned long ex_xprtsec_modes; struct export_stats *ex_stats; }; @@ -92,7 +93,7 @@ struct svc_expkey { u32 ek_fsid[6]; struct path ek_path; - struct rcu_head ek_rcu; + struct rcu_work ek_rwork; }; #define EX_ISSYNC(exp) (!((exp)->ex_flags & NFSEXP_ASYNC)) @@ -110,6 +111,8 @@ __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp, /* * Function declarations */ +int nfsd_export_wq_init(void); +void nfsd_export_wq_shutdown(void); int nfsd_export_init(struct net *); void nfsd_export_shutdown(struct net *); void nfsd_export_flush(struct net *); diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 41dfba5ab8b8..9d234913100b 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -6281,9 +6281,14 @@ nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) int len = xdr->buf->len - (op_status_offset + XDR_UNIT); so->so_replay.rp_status = op->status; - so->so_replay.rp_buflen = len; - read_bytes_from_xdr_buf(xdr->buf, op_status_offset + XDR_UNIT, + if (len <= NFSD4_REPLAY_ISIZE) { + so->so_replay.rp_buflen = len; + read_bytes_from_xdr_buf(xdr->buf, + op_status_offset + XDR_UNIT, so->so_replay.rp_buf, len); + } else { + so->so_replay.rp_buflen = 0; + } } status: op->status = nfsd4_map_status(op->status, diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index e9acd2cd602c..71aabdaa1d15 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -149,9 +149,19 @@ static int exports_net_open(struct net *net, struct file *file) seq = file->private_data; seq->private = nn->svc_export_cache; + get_net(net); return 0; } +static int exports_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct cache_detail *cd = seq->private; + + put_net(cd->net); + return seq_release(inode, file); +} + static int exports_nfsd_open(struct inode *inode, struct file *file) { return exports_net_open(inode->i_sb->s_fs_info, file); @@ -161,7 +171,7 @@ static const struct file_operations exports_nfsd_operations = { .open = exports_nfsd_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = exports_release, }; static int export_features_show(struct seq_file *m, void *v) @@ -377,15 +387,15 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) } /* - * write_threads - Start NFSD, or report the current number of running threads + * write_threads - Start NFSD, or report the configured number of threads * * Input: * buf: ignored * size: zero * Output: * On success: passed-in buffer filled with '\n'-terminated C - * string numeric value representing the number of - * running NFSD threads; + * string numeric value representing the configured + * number of NFSD threads; * return code is the size in bytes of the string * On error: return code is zero * @@ -399,8 +409,8 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) * Output: * On success: NFS service is started; * passed-in buffer filled with '\n'-terminated C - * string numeric value representing the number of - * running NFSD threads; + * string numeric value representing the configured + * number of NFSD threads; * return code is the size in bytes of the string * On error: return code is zero or a negative errno value */ @@ -430,7 +440,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) } /* - * write_pool_threads - Set or report the current number of threads per pool + * write_pool_threads - Set or report the configured number of threads per pool * * Input: * buf: ignored @@ -447,7 +457,7 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) * Output: * On success: passed-in buffer filled with '\n'-terminated C * string containing integer values representing the - * number of NFSD threads in each pool; + * configured number of NFSD threads in each pool; * return code is the size in bytes of the string * On error: return code is zero or a negative errno value */ @@ -1376,7 +1386,7 @@ static const struct proc_ops exports_proc_ops = { .proc_open = exports_proc_open, .proc_read = seq_read, .proc_lseek = seq_lseek, - .proc_release = seq_release, + .proc_release = exports_release, }; static int create_proc_exports_entry(void) @@ -1647,7 +1657,7 @@ int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info) if (attr) nn->min_threads = nla_get_u32(attr); - ret = nfsd_svc(nrpools, nthreads, net, get_current_cred(), scope); + ret = nfsd_svc(nrpools, nthreads, net, current_cred(), scope); if (ret > 0) ret = 0; out_unlock: @@ -1657,7 +1667,7 @@ out_unlock: } /** - * nfsd_nl_threads_get_doit - get the number of running threads + * nfsd_nl_threads_get_doit - get the maximum number of running threads * @skb: reply buffer * @info: netlink metadata and command arguments * @@ -1700,7 +1710,7 @@ int nfsd_nl_threads_get_doit(struct sk_buff *skb, struct genl_info *info) struct svc_pool *sp = &nn->nfsd_serv->sv_pools[i]; err = nla_put_u32(skb, NFSD_A_SERVER_THREADS, - sp->sp_nrthreads); + sp->sp_nrthrmax); if (err) goto err_unlock; } @@ -2000,7 +2010,7 @@ int nfsd_nl_listener_set_doit(struct sk_buff *skb, struct genl_info *info) } ret = svc_xprt_create_from_sa(serv, xcl_name, net, sa, 0, - get_current_cred()); + current_cred()); /* always save the latest error */ if (ret < 0) err = ret; @@ -2259,9 +2269,12 @@ static int __init init_nfsd(void) if (retval) goto out_free_pnfs; nfsd_lockd_init(); /* lockd->nfsd callbacks */ + retval = nfsd_export_wq_init(); + if (retval) + goto out_free_lockd; retval = register_pernet_subsys(&nfsd_net_ops); if (retval < 0) - goto out_free_lockd; + goto out_free_export_wq; retval = register_cld_notifier(); if (retval) goto out_free_subsys; @@ -2290,6 +2303,8 @@ out_free_cld: unregister_cld_notifier(); out_free_subsys: unregister_pernet_subsys(&nfsd_net_ops); +out_free_export_wq: + nfsd_export_wq_shutdown(); out_free_lockd: nfsd_lockd_shutdown(); nfsd_drc_slab_free(); @@ -2310,6 +2325,7 @@ static void __exit exit_nfsd(void) nfsd4_destroy_laundry_wq(); unregister_cld_notifier(); unregister_pernet_subsys(&nfsd_net_ops); + nfsd_export_wq_shutdown(); nfsd_drc_slab_free(); nfsd_lockd_shutdown(); nfsd4_free_slabs(); diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 0887ee601d3c..4a04208393b8 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -239,12 +239,13 @@ static void nfsd_net_free(struct percpu_ref *ref) int nfsd_nrthreads(struct net *net) { - int rv = 0; + int i, rv = 0; struct nfsd_net *nn = net_generic(net, nfsd_net_id); mutex_lock(&nfsd_mutex); if (nn->nfsd_serv) - rv = nn->nfsd_serv->sv_nrthreads; + for (i = 0; i < nn->nfsd_serv->sv_nrpools; ++i) + rv += nn->nfsd_serv->sv_pools[i].sp_nrthrmax; mutex_unlock(&nfsd_mutex); return rv; } @@ -659,7 +660,7 @@ int nfsd_get_nrthreads(int n, int *nthreads, struct net *net) if (serv) for (i = 0; i < serv->sv_nrpools && i < n; i++) - nthreads[i] = serv->sv_pools[i].sp_nrthreads; + nthreads[i] = serv->sv_pools[i].sp_nrthrmax; return 0; } diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 6fcbf1e427d4..c0ca115c3b74 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -541,11 +541,18 @@ struct nfs4_client_reclaim { struct xdr_netobj cr_princhash; }; -/* A reasonable value for REPLAY_ISIZE was estimated as follows: - * The OPEN response, typically the largest, requires - * 4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) + 8(verifier) + - * 4(deleg. type) + 8(deleg. stateid) + 4(deleg. recall flag) + - * 20(deleg. space limit) + ~32(deleg. ace) = 112 bytes +/* + * REPLAY_ISIZE is sized for an OPEN response with delegation: + * 4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) + + * 8(verifier) + 4(deleg. type) + 8(deleg. stateid) + + * 4(deleg. recall flag) + 20(deleg. space limit) + + * ~32(deleg. ace) = 112 bytes + * + * Some responses can exceed this. A LOCK denial includes the conflicting + * lock owner, which can be up to 1024 bytes (NFS4_OPAQUE_LIMIT). Responses + * larger than REPLAY_ISIZE are not cached in rp_ibuf; only rp_status is + * saved. Enlarging this constant increases the size of every + * nfs4_stateowner. */ #define NFSD4_REPLAY_ISIZE 112 diff --git a/fs/nsfs.c b/fs/nsfs.c index db91de208645..c215878d55e8 100644 --- a/fs/nsfs.c +++ b/fs/nsfs.c @@ -199,6 +199,17 @@ static bool nsfs_ioctl_valid(unsigned int cmd) return false; } +static bool may_use_nsfs_ioctl(unsigned int cmd) +{ + switch (_IOC_NR(cmd)) { + case _IOC_NR(NS_MNT_GET_NEXT): + fallthrough; + case _IOC_NR(NS_MNT_GET_PREV): + return may_see_all_namespaces(); + } + return true; +} + static long ns_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) { @@ -214,6 +225,8 @@ static long ns_ioctl(struct file *filp, unsigned int ioctl, if (!nsfs_ioctl_valid(ioctl)) return -ENOIOCTLCMD; + if (!may_use_nsfs_ioctl(ioctl)) + return -EPERM; ns = get_proc_ns(file_inode(filp)); switch (ioctl) { @@ -614,7 +627,7 @@ static struct dentry *nsfs_fh_to_dentry(struct super_block *sb, struct fid *fh, return ERR_PTR(-EOPNOTSUPP); } - if (owning_ns && !ns_capable(owning_ns, CAP_SYS_ADMIN)) { + if (owning_ns && !may_see_all_namespaces()) { ns->ops->put(ns); return ERR_PTR(-EPERM); } diff --git a/fs/pidfs.c b/fs/pidfs.c index 318253344b5c..e3825ee246be 100644 --- a/fs/pidfs.c +++ b/fs/pidfs.c @@ -608,9 +608,8 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct user_namespace *user_ns; user_ns = task_cred_xxx(task, user_ns); - if (!ns_ref_get(user_ns)) - break; - ns_common = to_ns_common(user_ns); + if (ns_ref_get(user_ns)) + ns_common = to_ns_common(user_ns); } #endif break; @@ -620,9 +619,8 @@ static long pidfd_ioctl(struct file *file, unsigned int cmd, unsigned long arg) struct pid_namespace *pid_ns; pid_ns = task_active_pid_ns(task); - if (!ns_ref_get(pid_ns)) - break; - ns_common = to_ns_common(pid_ns); + if (ns_ref_get(pid_ns)) + ns_common = to_ns_common(pid_ns); } #endif break; diff --git a/fs/proc/base.c b/fs/proc/base.c index 4eec684baca9..4c863d17dfb4 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2128,6 +2128,9 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx, ino_t ino = 1; child = try_lookup_noperm(&qname, dir); + if (IS_ERR(child)) + goto end_instantiate; + if (!child) { DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); child = d_alloc_parallel(dir, &qname, &wq); diff --git a/fs/smb/client/Makefile b/fs/smb/client/Makefile index 3abd357d6df6..26b6105f04d1 100644 --- a/fs/smb/client/Makefile +++ b/fs/smb/client/Makefile @@ -56,4 +56,6 @@ $(obj)/smb2maperror.o: $(obj)/smb2_mapping_table.c quiet_cmd_gen_smb2_mapping = GEN $@ cmd_gen_smb2_mapping = perl $(src)/gen_smb2_mapping $< $@ +obj-$(CONFIG_SMB_KUNIT_TESTS) += smb2maperror_test.o + clean-files += smb2_mapping_table.c diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c index c327c246a9b4..04bb95091f49 100644 --- a/fs/smb/client/cached_dir.c +++ b/fs/smb/client/cached_dir.c @@ -118,7 +118,7 @@ static const char *path_no_prefix(struct cifs_sb_info *cifs_sb, if (!*path) return path; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) && + if ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_USE_PREFIX_PATH) && cifs_sb->prepath) { len = strlen(cifs_sb->prepath) + 1; if (unlikely(len > strlen(path))) diff --git a/fs/smb/client/cifs_fs_sb.h b/fs/smb/client/cifs_fs_sb.h index 5e8d163cb5f8..84e7e366b0ff 100644 --- a/fs/smb/client/cifs_fs_sb.h +++ b/fs/smb/client/cifs_fs_sb.h @@ -55,7 +55,7 @@ struct cifs_sb_info { struct nls_table *local_nls; struct smb3_fs_context *ctx; atomic_t active; - unsigned int mnt_cifs_flags; + atomic_t mnt_cifs_flags; struct delayed_work prune_tlinks; struct rcu_head rcu; diff --git a/fs/smb/client/cifs_ioctl.h b/fs/smb/client/cifs_ioctl.h index b51ce64fcccf..147496ac9f9f 100644 --- a/fs/smb/client/cifs_ioctl.h +++ b/fs/smb/client/cifs_ioctl.h @@ -122,11 +122,3 @@ struct smb3_notify_info { #define CIFS_GOING_FLAGS_DEFAULT 0x0 /* going down */ #define CIFS_GOING_FLAGS_LOGFLUSH 0x1 /* flush log but not data */ #define CIFS_GOING_FLAGS_NOLOGFLUSH 0x2 /* don't flush log nor data */ - -static inline bool cifs_forced_shutdown(struct cifs_sb_info *sbi) -{ - if (CIFS_MOUNT_SHUTDOWN & sbi->mnt_cifs_flags) - return true; - else - return false; -} diff --git a/fs/smb/client/cifs_unicode.c b/fs/smb/client/cifs_unicode.c index e7891b4406f2..e2edc207cef2 100644 --- a/fs/smb/client/cifs_unicode.c +++ b/fs/smb/client/cifs_unicode.c @@ -11,20 +11,6 @@ #include "cifsglob.h" #include "cifs_debug.h" -int cifs_remap(struct cifs_sb_info *cifs_sb) -{ - int map_type; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR) - map_type = SFM_MAP_UNI_RSVD; - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR) - map_type = SFU_MAP_UNI_RSVD; - else - map_type = NO_MAP_UNI_RSVD; - - return map_type; -} - /* Convert character using the SFU - "Services for Unix" remapping range */ static bool convert_sfu_char(const __u16 src_char, char *target) diff --git a/fs/smb/client/cifs_unicode.h b/fs/smb/client/cifs_unicode.h index 9249db3b78c3..3e9cd9acf0a9 100644 --- a/fs/smb/client/cifs_unicode.h +++ b/fs/smb/client/cifs_unicode.h @@ -22,6 +22,7 @@ #include <linux/types.h> #include <linux/nls.h> #include "../../nls/nls_ucs2_utils.h" +#include "cifsglob.h" /* * Macs use an older "SFM" mapping of the symbols above. Fortunately it does @@ -65,10 +66,21 @@ char *cifs_strndup_from_utf16(const char *src, const int maxlen, const struct nls_table *codepage); int cifsConvertToUTF16(__le16 *target, const char *source, int srclen, const struct nls_table *cp, int map_chars); -int cifs_remap(struct cifs_sb_info *cifs_sb); __le16 *cifs_strndup_to_utf16(const char *src, const int maxlen, int *utf16_len, const struct nls_table *cp, int remap); wchar_t cifs_toupper(wchar_t in); +static inline int cifs_remap(const struct cifs_sb_info *cifs_sb) +{ + unsigned int sbflags = cifs_sb_flags(cifs_sb); + + if (sbflags & CIFS_MOUNT_MAP_SFM_CHR) + return SFM_MAP_UNI_RSVD; + if (sbflags & CIFS_MOUNT_MAP_SPECIAL_CHR) + return SFU_MAP_UNI_RSVD; + + return NO_MAP_UNI_RSVD; +} + #endif /* _CIFS_UNICODE_H */ diff --git a/fs/smb/client/cifsacl.c b/fs/smb/client/cifsacl.c index 6fa12c901c14..c920039d733c 100644 --- a/fs/smb/client/cifsacl.c +++ b/fs/smb/client/cifsacl.c @@ -356,7 +356,7 @@ sid_to_id(struct cifs_sb_info *cifs_sb, struct smb_sid *psid, psid->num_subauth, SID_MAX_SUB_AUTHORITIES); } - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) || + if ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_UID_FROM_ACL) || (cifs_sb_master_tcon(cifs_sb)->posix_extensions)) { uint32_t unix_id; bool is_group; @@ -1489,7 +1489,7 @@ struct smb_ntsd *get_cifs_acl(struct cifs_sb_info *cifs_sb, struct cifsFileInfo *open_file = NULL; if (inode) - open_file = find_readable_file(CIFS_I(inode), true); + open_file = find_readable_file(CIFS_I(inode), FIND_FSUID_ONLY); if (!open_file) return get_cifs_acl_by_path(cifs_sb, path, pacllen, info); @@ -1612,7 +1612,8 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, struct smb_acl *dacl_ptr = NULL; struct smb_ntsd *pntsd = NULL; /* acl obtained from server */ struct smb_ntsd *pnntsd = NULL; /* modified acl to be sent to server */ - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + unsigned int sbflags; struct tcon_link *tlink; struct smb_version_operations *ops; bool mode_from_sid, id_from_sid; @@ -1643,15 +1644,9 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, return rc; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) - mode_from_sid = true; - else - mode_from_sid = false; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) - id_from_sid = true; - else - id_from_sid = false; + sbflags = cifs_sb_flags(cifs_sb); + mode_from_sid = sbflags & CIFS_MOUNT_MODE_FROM_SID; + id_from_sid = sbflags & CIFS_MOUNT_UID_FROM_ACL; /* Potentially, five new ACEs can be added to the ACL for U,G,O mapping */ if (pnmode && *pnmode != NO_CHANGE_64) { /* chmod */ diff --git a/fs/smb/client/cifsfs.c b/fs/smb/client/cifsfs.c index 99b04234a08e..32d0305a1239 100644 --- a/fs/smb/client/cifsfs.c +++ b/fs/smb/client/cifsfs.c @@ -226,16 +226,18 @@ cifs_sb_deactive(struct super_block *sb) static int cifs_read_super(struct super_block *sb) { - struct inode *inode; struct cifs_sb_info *cifs_sb; struct cifs_tcon *tcon; + unsigned int sbflags; struct timespec64 ts; + struct inode *inode; int rc = 0; cifs_sb = CIFS_SB(sb); tcon = cifs_sb_master_tcon(cifs_sb); + sbflags = cifs_sb_flags(cifs_sb); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIXACL) + if (sbflags & CIFS_MOUNT_POSIXACL) sb->s_flags |= SB_POSIXACL; if (tcon->snapshot_time) @@ -311,7 +313,7 @@ cifs_read_super(struct super_block *sb) } #ifdef CONFIG_CIFS_NFSD_EXPORT - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) { + if (sbflags & CIFS_MOUNT_SERVER_INUM) { cifs_dbg(FYI, "export ops supported\n"); sb->s_export_op = &cifs_export_ops; } @@ -330,10 +332,14 @@ static void cifs_kill_sb(struct super_block *sb) /* * We need to release all dentries for the cached directories - * before we kill the sb. + * and close all deferred file handles before we kill the sb. */ if (cifs_sb->root) { close_all_cached_dirs(cifs_sb); + cifs_close_all_deferred_files_sb(cifs_sb); + + /* Wait for all pending oplock breaks to complete */ + flush_workqueue(cifsoplockd_wq); /* finally release root dentry */ dput(cifs_sb->root); @@ -389,8 +395,7 @@ statfs_out: static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len) { - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file); - struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + struct cifs_tcon *tcon = cifs_sb_master_tcon(CIFS_SB(file)); struct TCP_Server_Info *server = tcon->ses->server; struct inode *inode = file_inode(file); int rc; @@ -418,11 +423,9 @@ out_unlock: static int cifs_permission(struct mnt_idmap *idmap, struct inode *inode, int mask) { - struct cifs_sb_info *cifs_sb; + unsigned int sbflags = cifs_sb_flags(CIFS_SB(inode)); - cifs_sb = CIFS_SB(inode->i_sb); - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) { + if (sbflags & CIFS_MOUNT_NO_PERM) { if ((mask & MAY_EXEC) && !execute_ok(inode)) return -EACCES; else @@ -568,15 +571,17 @@ cifs_show_security(struct seq_file *s, struct cifs_ses *ses) static void cifs_show_cache_flavor(struct seq_file *s, struct cifs_sb_info *cifs_sb) { + unsigned int sbflags = cifs_sb_flags(cifs_sb); + seq_puts(s, ",cache="); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) + if (sbflags & CIFS_MOUNT_STRICT_IO) seq_puts(s, "strict"); - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) + else if (sbflags & CIFS_MOUNT_DIRECT_IO) seq_puts(s, "none"); - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE) + else if (sbflags & CIFS_MOUNT_RW_CACHE) seq_puts(s, "singleclient"); /* assume only one client access */ - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE) + else if (sbflags & CIFS_MOUNT_RO_CACHE) seq_puts(s, "ro"); /* read only caching assumed */ else seq_puts(s, "loose"); @@ -637,6 +642,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root) struct cifs_sb_info *cifs_sb = CIFS_SB(root->d_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct sockaddr *srcaddr; + unsigned int sbflags; + srcaddr = (struct sockaddr *)&tcon->ses->server->srcaddr; seq_show_option(s, "vers", tcon->ses->server->vals->version_string); @@ -670,16 +677,17 @@ cifs_show_options(struct seq_file *s, struct dentry *root) (int)(srcaddr->sa_family)); } + sbflags = cifs_sb_flags(cifs_sb); seq_printf(s, ",uid=%u", from_kuid_munged(&init_user_ns, cifs_sb->ctx->linux_uid)); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID) + if (sbflags & CIFS_MOUNT_OVERR_UID) seq_puts(s, ",forceuid"); else seq_puts(s, ",noforceuid"); seq_printf(s, ",gid=%u", from_kgid_munged(&init_user_ns, cifs_sb->ctx->linux_gid)); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID) + if (sbflags & CIFS_MOUNT_OVERR_GID) seq_puts(s, ",forcegid"); else seq_puts(s, ",noforcegid"); @@ -722,53 +730,53 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_puts(s, ",unix"); else seq_puts(s, ",nounix"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) + if (sbflags & CIFS_MOUNT_NO_DFS) seq_puts(s, ",nodfs"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) + if (sbflags & CIFS_MOUNT_POSIX_PATHS) seq_puts(s, ",posixpaths"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) + if (sbflags & CIFS_MOUNT_SET_UID) seq_puts(s, ",setuids"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) + if (sbflags & CIFS_MOUNT_UID_FROM_ACL) seq_puts(s, ",idsfromsid"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) + if (sbflags & CIFS_MOUNT_SERVER_INUM) seq_puts(s, ",serverino"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + if (sbflags & CIFS_MOUNT_RWPIDFORWARD) seq_puts(s, ",rwpidforward"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) + if (sbflags & CIFS_MOUNT_NOPOSIXBRL) seq_puts(s, ",forcemand"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + if (sbflags & CIFS_MOUNT_NO_XATTR) seq_puts(s, ",nouser_xattr"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR) + if (sbflags & CIFS_MOUNT_MAP_SPECIAL_CHR) seq_puts(s, ",mapchars"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR) + if (sbflags & CIFS_MOUNT_MAP_SFM_CHR) seq_puts(s, ",mapposix"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) + if (sbflags & CIFS_MOUNT_UNX_EMUL) seq_puts(s, ",sfu"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + if (sbflags & CIFS_MOUNT_NO_BRL) seq_puts(s, ",nobrl"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_HANDLE_CACHE) + if (sbflags & CIFS_MOUNT_NO_HANDLE_CACHE) seq_puts(s, ",nohandlecache"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) + if (sbflags & CIFS_MOUNT_MODE_FROM_SID) seq_puts(s, ",modefromsid"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) + if (sbflags & CIFS_MOUNT_CIFS_ACL) seq_puts(s, ",cifsacl"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) + if (sbflags & CIFS_MOUNT_DYNPERM) seq_puts(s, ",dynperm"); if (root->d_sb->s_flags & SB_POSIXACL) seq_puts(s, ",acl"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) + if (sbflags & CIFS_MOUNT_MF_SYMLINKS) seq_puts(s, ",mfsymlinks"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) + if (sbflags & CIFS_MOUNT_FSCACHE) seq_puts(s, ",fsc"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC) + if (sbflags & CIFS_MOUNT_NOSSYNC) seq_puts(s, ",nostrictsync"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) + if (sbflags & CIFS_MOUNT_NO_PERM) seq_puts(s, ",noperm"); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_BACKUPUID) + if (sbflags & CIFS_MOUNT_CIFS_BACKUPUID) seq_printf(s, ",backupuid=%u", from_kuid_munged(&init_user_ns, cifs_sb->ctx->backupuid)); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_BACKUPGID) + if (sbflags & CIFS_MOUNT_CIFS_BACKUPGID) seq_printf(s, ",backupgid=%u", from_kgid_munged(&init_user_ns, cifs_sb->ctx->backupgid)); @@ -864,7 +872,6 @@ static void cifs_umount_begin(struct super_block *sb) spin_unlock(&tcon->tc_lock); spin_unlock(&cifs_tcp_ses_lock); - cifs_close_all_deferred_files(tcon); /* cancel_brl_requests(tcon); */ /* BB mark all brl mids as exiting */ /* cancel_notify_requests(tcon); */ if (tcon->ses && tcon->ses->server) { @@ -909,10 +916,10 @@ static int cifs_write_inode(struct inode *inode, struct writeback_control *wbc) static int cifs_drop_inode(struct inode *inode) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + unsigned int sbflags = cifs_sb_flags(CIFS_SB(inode)); /* no serverino => unconditional eviction */ - return !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) || + return !(sbflags & CIFS_MOUNT_SERVER_INUM) || inode_generic_drop(inode); } @@ -950,7 +957,7 @@ cifs_get_root(struct smb3_fs_context *ctx, struct super_block *sb) char *s, *p; char sep; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_USE_PREFIX_PATH) return dget(sb->s_root); full_path = cifs_build_path_to_root(ctx, cifs_sb, @@ -1262,7 +1269,7 @@ static int cifs_precopy_set_eof(struct inode *src_inode, struct cifsInodeInfo *s struct cifsFileInfo *writeable_srcfile; int rc = -EINVAL; - writeable_srcfile = find_writable_file(src_cifsi, FIND_WR_FSUID_ONLY); + writeable_srcfile = find_writable_file(src_cifsi, FIND_FSUID_ONLY); if (writeable_srcfile) { if (src_tcon->ses->server->ops->set_file_size) rc = src_tcon->ses->server->ops->set_file_size( diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 080ea601c209..709e96e07791 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -20,6 +20,7 @@ #include <linux/utsname.h> #include <linux/sched/mm.h> #include <linux/netfs.h> +#include <linux/fcntl.h> #include "cifs_fs_sb.h" #include "cifsacl.h" #include <crypto/internal/hash.h> @@ -1580,24 +1581,59 @@ CIFS_I(struct inode *inode) return container_of(inode, struct cifsInodeInfo, netfs.inode); } -static inline struct cifs_sb_info * -CIFS_SB(struct super_block *sb) +static inline void *cinode_to_fsinfo(struct cifsInodeInfo *cinode) +{ + return cinode->netfs.inode.i_sb->s_fs_info; +} + +static inline void *super_to_fsinfo(struct super_block *sb) { return sb->s_fs_info; } -static inline struct cifs_sb_info * -CIFS_FILE_SB(struct file *file) +static inline void *inode_to_fsinfo(struct inode *inode) +{ + return inode->i_sb->s_fs_info; +} + +static inline void *file_to_fsinfo(struct file *file) +{ + return file_inode(file)->i_sb->s_fs_info; +} + +static inline void *dentry_to_fsinfo(struct dentry *dentry) +{ + return dentry->d_sb->s_fs_info; +} + +static inline void *const_dentry_to_fsinfo(const struct dentry *dentry) +{ + return dentry->d_sb->s_fs_info; +} + +#define CIFS_SB(_ptr) \ + ((struct cifs_sb_info *) \ + _Generic((_ptr), \ + struct cifsInodeInfo * : cinode_to_fsinfo, \ + const struct dentry * : const_dentry_to_fsinfo, \ + struct super_block * : super_to_fsinfo, \ + struct dentry * : dentry_to_fsinfo, \ + struct inode * : inode_to_fsinfo, \ + struct file * : file_to_fsinfo)(_ptr)) + +/* + * Use atomic_t for @cifs_sb->mnt_cifs_flags as it is currently accessed + * locklessly and may be changed concurrently by mount/remount and reconnect + * paths. + */ +static inline unsigned int cifs_sb_flags(const struct cifs_sb_info *cifs_sb) { - return CIFS_SB(file_inode(file)->i_sb); + return atomic_read(&cifs_sb->mnt_cifs_flags); } static inline char CIFS_DIR_SEP(const struct cifs_sb_info *cifs_sb) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) - return '/'; - else - return '\\'; + return (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_POSIX_PATHS) ? '/' : '\\'; } static inline void @@ -1849,12 +1885,12 @@ static inline bool is_replayable_error(int error) } -/* cifs_get_writable_file() flags */ -enum cifs_writable_file_flags { - FIND_WR_ANY = 0U, - FIND_WR_FSUID_ONLY = (1U << 0), - FIND_WR_WITH_DELETE = (1U << 1), - FIND_WR_NO_PENDING_DELETE = (1U << 2), +enum cifs_find_flags { + FIND_ANY = 0U, + FIND_FSUID_ONLY = (1U << 0), + FIND_WITH_DELETE = (1U << 1), + FIND_NO_PENDING_DELETE = (1U << 2), + FIND_OPEN_FLAGS = (1U << 3), }; #define MID_FREE 0 @@ -2314,9 +2350,8 @@ static inline bool __cifs_cache_state_check(struct cifsInodeInfo *cinode, unsigned int oplock_flags, unsigned int sb_flags) { - struct cifs_sb_info *cifs_sb = CIFS_SB(cinode->netfs.inode.i_sb); + unsigned int sflags = cifs_sb_flags(CIFS_SB(cinode)); unsigned int oplock = READ_ONCE(cinode->oplock); - unsigned int sflags = cifs_sb->mnt_cifs_flags; return (oplock & oplock_flags) || (sflags & sb_flags); } @@ -2336,4 +2371,25 @@ static inline void cifs_reset_oplock(struct cifsInodeInfo *cinode) WRITE_ONCE(cinode->oplock, 0); } +static inline bool cifs_forced_shutdown(const struct cifs_sb_info *sbi) +{ + return cifs_sb_flags(sbi) & CIFS_MOUNT_SHUTDOWN; +} + +static inline int cifs_open_create_options(unsigned int oflags, int opts) +{ + /* O_SYNC also has bit for O_DSYNC so following check picks up either */ + if (oflags & O_SYNC) + opts |= CREATE_WRITE_THROUGH; + if (oflags & O_DIRECT) + opts |= CREATE_NO_BUFFER; + return opts; +} + +/* + * The number of blocks is not related to (i_size / i_blksize), but instead + * 512 byte (2**9) size is required for calculating num blocks. + */ +#define CIFS_INO_BLOCKS(size) DIV_ROUND_UP_ULL((u64)(size), 512) + #endif /* _CIFS_GLOB_H */ diff --git a/fs/smb/client/cifsproto.h b/fs/smb/client/cifsproto.h index 96d6b5325aa3..884bfa1cf0b4 100644 --- a/fs/smb/client/cifsproto.h +++ b/fs/smb/client/cifsproto.h @@ -138,12 +138,14 @@ void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t result); struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode, int flags); -int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, - struct cifsFileInfo **ret_file); +int __cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, + unsigned int find_flags, unsigned int open_flags, + struct cifsFileInfo **ret_file); int cifs_get_writable_path(struct cifs_tcon *tcon, const char *name, int flags, struct cifsFileInfo **ret_file); -struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, - bool fsuid_only); +struct cifsFileInfo *__find_readable_file(struct cifsInodeInfo *cifs_inode, + unsigned int find_flags, + unsigned int open_flags); int cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, struct cifsFileInfo **ret_file); int cifs_get_hardlink_path(struct cifs_tcon *tcon, struct inode *inode, @@ -261,6 +263,7 @@ void cifs_close_deferred_file(struct cifsInodeInfo *cifs_inode); void cifs_close_all_deferred_files(struct cifs_tcon *tcon); +void cifs_close_all_deferred_files_sb(struct cifs_sb_info *cifs_sb); void cifs_close_deferred_file_under_dentry(struct cifs_tcon *tcon, struct dentry *dentry); @@ -595,4 +598,20 @@ static inline void cifs_sg_set_buf(struct sg_table *sgtable, } } +static inline int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, + unsigned int find_flags, + struct cifsFileInfo **ret_file) +{ + find_flags &= ~FIND_OPEN_FLAGS; + return __cifs_get_writable_file(cifs_inode, find_flags, 0, ret_file); +} + +static inline struct cifsFileInfo * +find_readable_file(struct cifsInodeInfo *cinode, unsigned int find_flags) +{ + find_flags &= ~FIND_OPEN_FLAGS; + find_flags |= FIND_NO_PENDING_DELETE; + return __find_readable_file(cinode, find_flags, 0); +} + #endif /* _CIFSPROTO_H */ diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 33dfe116ca52..69b38f0ccf2b 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -1955,6 +1955,10 @@ static int match_session(struct cifs_ses *ses, case Kerberos: if (!uid_eq(ctx->cred_uid, ses->cred_uid)) return 0; + if (strncmp(ses->user_name ?: "", + ctx->username ?: "", + CIFS_MAX_USERNAME_LEN)) + return 0; break; case NTLMv2: case RawNTLMSSP: @@ -2167,9 +2171,6 @@ void __cifs_put_smb_ses(struct cifs_ses *ses) #ifdef CONFIG_KEYS -/* strlen("cifs:a:") + CIFS_MAX_DOMAINNAME_LEN + 1 */ -#define CIFSCREDS_DESC_SIZE (7 + CIFS_MAX_DOMAINNAME_LEN + 1) - /* Populate username and pw fields from keyring if possible */ static int cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) @@ -2177,6 +2178,7 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) int rc = 0; int is_domain = 0; const char *delim, *payload; + size_t desc_sz; char *desc; ssize_t len; struct key *key; @@ -2185,7 +2187,9 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) struct sockaddr_in6 *sa6; const struct user_key_payload *upayload; - desc = kmalloc(CIFSCREDS_DESC_SIZE, GFP_KERNEL); + /* "cifs:a:" and "cifs:d:" are the same length; +1 for NUL terminator */ + desc_sz = strlen("cifs:a:") + CIFS_MAX_DOMAINNAME_LEN + 1; + desc = kmalloc(desc_sz, GFP_KERNEL); if (!desc) return -ENOMEM; @@ -2193,11 +2197,11 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) switch (server->dstaddr.ss_family) { case AF_INET: sa = (struct sockaddr_in *)&server->dstaddr; - sprintf(desc, "cifs:a:%pI4", &sa->sin_addr.s_addr); + snprintf(desc, desc_sz, "cifs:a:%pI4", &sa->sin_addr.s_addr); break; case AF_INET6: sa6 = (struct sockaddr_in6 *)&server->dstaddr; - sprintf(desc, "cifs:a:%pI6c", &sa6->sin6_addr.s6_addr); + snprintf(desc, desc_sz, "cifs:a:%pI6c", &sa6->sin6_addr.s6_addr); break; default: cifs_dbg(FYI, "Bad ss_family (%hu)\n", @@ -2216,7 +2220,7 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) } /* didn't work, try to find a domain key */ - sprintf(desc, "cifs:d:%s", ses->domainName); + snprintf(desc, desc_sz, "cifs:d:%s", ses->domainName); cifs_dbg(FYI, "%s: desc=%s\n", __func__, desc); key = request_key(&key_type_logon, desc, ""); if (IS_ERR(key)) { @@ -2236,7 +2240,6 @@ cifs_set_cifscreds(struct smb3_fs_context *ctx, struct cifs_ses *ses) /* find first : in payload */ payload = upayload->data; delim = strnchr(payload, upayload->datalen, ':'); - cifs_dbg(FYI, "payload=%s\n", payload); if (!delim) { cifs_dbg(FYI, "Unable to find ':' in payload (datalen=%d)\n", upayload->datalen); @@ -2915,8 +2918,8 @@ compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data) { struct cifs_sb_info *old = CIFS_SB(sb); struct cifs_sb_info *new = mnt_data->cifs_sb; - unsigned int oldflags = old->mnt_cifs_flags & CIFS_MOUNT_MASK; - unsigned int newflags = new->mnt_cifs_flags & CIFS_MOUNT_MASK; + unsigned int oldflags = cifs_sb_flags(old) & CIFS_MOUNT_MASK; + unsigned int newflags = cifs_sb_flags(new) & CIFS_MOUNT_MASK; if ((sb->s_flags & CIFS_MS_MASK) != (mnt_data->flags & CIFS_MS_MASK)) return 0; @@ -2971,9 +2974,9 @@ static int match_prepath(struct super_block *sb, struct smb3_fs_context *ctx = mnt_data->ctx; struct cifs_sb_info *old = CIFS_SB(sb); struct cifs_sb_info *new = mnt_data->cifs_sb; - bool old_set = (old->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) && + bool old_set = (cifs_sb_flags(old) & CIFS_MOUNT_USE_PREFIX_PATH) && old->prepath; - bool new_set = (new->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) && + bool new_set = (cifs_sb_flags(new) & CIFS_MOUNT_USE_PREFIX_PATH) && new->prepath; if (tcon->origin_fullpath && @@ -3004,7 +3007,7 @@ cifs_match_super(struct super_block *sb, void *data) cifs_sb = CIFS_SB(sb); /* We do not want to use a superblock that has been shutdown */ - if (CIFS_MOUNT_SHUTDOWN & cifs_sb->mnt_cifs_flags) { + if (cifs_forced_shutdown(cifs_sb)) { spin_unlock(&cifs_tcp_ses_lock); return 0; } @@ -3469,6 +3472,8 @@ ip_connect(struct TCP_Server_Info *server) int cifs_setup_cifs_sb(struct cifs_sb_info *cifs_sb) { struct smb3_fs_context *ctx = cifs_sb->ctx; + unsigned int sbflags; + int rc = 0; INIT_DELAYED_WORK(&cifs_sb->prune_tlinks, cifs_prune_tlinks); INIT_LIST_HEAD(&cifs_sb->tcon_sb_link); @@ -3493,17 +3498,16 @@ int cifs_setup_cifs_sb(struct cifs_sb_info *cifs_sb) } ctx->local_nls = cifs_sb->local_nls; - smb3_update_mnt_flags(cifs_sb); + sbflags = smb3_update_mnt_flags(cifs_sb); if (ctx->direct_io) cifs_dbg(FYI, "mounting share using direct i/o\n"); if (ctx->cache_ro) { cifs_dbg(VFS, "mounting share with read only caching. Ensure that the share will not be modified while in use.\n"); - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_RO_CACHE; + sbflags |= CIFS_MOUNT_RO_CACHE; } else if (ctx->cache_rw) { cifs_dbg(VFS, "mounting share in single client RW caching mode. Ensure that no other systems will be accessing the share.\n"); - cifs_sb->mnt_cifs_flags |= (CIFS_MOUNT_RO_CACHE | - CIFS_MOUNT_RW_CACHE); + sbflags |= CIFS_MOUNT_RO_CACHE | CIFS_MOUNT_RW_CACHE; } if ((ctx->cifs_acl) && (ctx->dynperm)) @@ -3512,16 +3516,19 @@ int cifs_setup_cifs_sb(struct cifs_sb_info *cifs_sb) if (ctx->prepath) { cifs_sb->prepath = kstrdup(ctx->prepath, GFP_KERNEL); if (cifs_sb->prepath == NULL) - return -ENOMEM; - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + rc = -ENOMEM; + else + sbflags |= CIFS_MOUNT_USE_PREFIX_PATH; } - return 0; + atomic_set(&cifs_sb->mnt_cifs_flags, sbflags); + return rc; } /* Release all succeed connections */ void cifs_mount_put_conns(struct cifs_mount_ctx *mnt_ctx) { + struct cifs_sb_info *cifs_sb = mnt_ctx->cifs_sb; int rc = 0; if (mnt_ctx->tcon) @@ -3533,7 +3540,7 @@ void cifs_mount_put_conns(struct cifs_mount_ctx *mnt_ctx) mnt_ctx->ses = NULL; mnt_ctx->tcon = NULL; mnt_ctx->server = NULL; - mnt_ctx->cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_POSIX_PATHS; + atomic_andnot(CIFS_MOUNT_POSIX_PATHS, &cifs_sb->mnt_cifs_flags); free_xid(mnt_ctx->xid); } @@ -3587,19 +3594,23 @@ out: int cifs_mount_get_tcon(struct cifs_mount_ctx *mnt_ctx) { struct TCP_Server_Info *server; + struct cifs_tcon *tcon = NULL; struct cifs_sb_info *cifs_sb; struct smb3_fs_context *ctx; - struct cifs_tcon *tcon = NULL; + unsigned int sbflags; int rc = 0; - if (WARN_ON_ONCE(!mnt_ctx || !mnt_ctx->server || !mnt_ctx->ses || !mnt_ctx->fs_ctx || - !mnt_ctx->cifs_sb)) { - rc = -EINVAL; - goto out; + if (WARN_ON_ONCE(!mnt_ctx)) + return -EINVAL; + if (WARN_ON_ONCE(!mnt_ctx->server || !mnt_ctx->ses || + !mnt_ctx->fs_ctx || !mnt_ctx->cifs_sb)) { + mnt_ctx->tcon = NULL; + return -EINVAL; } server = mnt_ctx->server; ctx = mnt_ctx->fs_ctx; cifs_sb = mnt_ctx->cifs_sb; + sbflags = cifs_sb_flags(cifs_sb); /* search for existing tcon to this server share */ tcon = cifs_get_tcon(mnt_ctx->ses, ctx); @@ -3614,9 +3625,9 @@ int cifs_mount_get_tcon(struct cifs_mount_ctx *mnt_ctx) * path (i.e., do not remap / and \ and do not map any special characters) */ if (tcon->posix_extensions) { - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; - cifs_sb->mnt_cifs_flags &= ~(CIFS_MOUNT_MAP_SFM_CHR | - CIFS_MOUNT_MAP_SPECIAL_CHR); + sbflags |= CIFS_MOUNT_POSIX_PATHS; + sbflags &= ~(CIFS_MOUNT_MAP_SFM_CHR | + CIFS_MOUNT_MAP_SPECIAL_CHR); } #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY @@ -3643,12 +3654,11 @@ int cifs_mount_get_tcon(struct cifs_mount_ctx *mnt_ctx) /* do not care if a following call succeed - informational */ if (!tcon->pipe && server->ops->qfs_tcon) { server->ops->qfs_tcon(mnt_ctx->xid, tcon, cifs_sb); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE) { + if (sbflags & CIFS_MOUNT_RO_CACHE) { if (tcon->fsDevInfo.DeviceCharacteristics & cpu_to_le32(FILE_READ_ONLY_DEVICE)) cifs_dbg(VFS, "mounted to read only share\n"); - else if ((cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_RW_CACHE) == 0) + else if (!(sbflags & CIFS_MOUNT_RW_CACHE)) cifs_dbg(VFS, "read only mount of RW share\n"); /* no need to log a RW mount of a typical RW share */ } @@ -3660,11 +3670,12 @@ int cifs_mount_get_tcon(struct cifs_mount_ctx *mnt_ctx) * Inside cifs_fscache_get_super_cookie it checks * that we do not get super cookie twice. */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) + if (sbflags & CIFS_MOUNT_FSCACHE) cifs_fscache_get_super_cookie(tcon); out: mnt_ctx->tcon = tcon; + atomic_set(&cifs_sb->mnt_cifs_flags, sbflags); return rc; } @@ -3783,7 +3794,8 @@ int cifs_is_path_remote(struct cifs_mount_ctx *mnt_ctx) cifs_sb, full_path, tcon->Flags & SMB_SHARE_IS_IN_DFS); if (rc != 0) { cifs_server_dbg(VFS, "cannot query dirs between root and final path, enabling CIFS_MOUNT_USE_PREFIX_PATH\n"); - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + atomic_or(CIFS_MOUNT_USE_PREFIX_PATH, + &cifs_sb->mnt_cifs_flags); rc = 0; } } @@ -3863,7 +3875,7 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) * Force the use of prefix path to support failover on DFS paths that resolve to targets * that have different prefix paths. */ - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + atomic_or(CIFS_MOUNT_USE_PREFIX_PATH, &cifs_sb->mnt_cifs_flags); kfree(cifs_sb->prepath); cifs_sb->prepath = ctx->prepath; ctx->prepath = NULL; @@ -4357,7 +4369,7 @@ cifs_sb_tlink(struct cifs_sb_info *cifs_sb) kuid_t fsuid = current_fsuid(); int err; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_MULTIUSER)) return cifs_get_tlink(cifs_sb_master_tlink(cifs_sb)); spin_lock(&cifs_sb->tlink_tree_lock); diff --git a/fs/smb/client/dfs_cache.c b/fs/smb/client/dfs_cache.c index 983132735d72..83f8cf2f8d2b 100644 --- a/fs/smb/client/dfs_cache.c +++ b/fs/smb/client/dfs_cache.c @@ -1333,7 +1333,7 @@ int dfs_cache_remount_fs(struct cifs_sb_info *cifs_sb) * Force the use of prefix path to support failover on DFS paths that resolve to targets * that have different prefix paths. */ - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + atomic_or(CIFS_MOUNT_USE_PREFIX_PATH, &cifs_sb->mnt_cifs_flags); refresh_tcon_referral(tcon, true); return 0; diff --git a/fs/smb/client/dir.c b/fs/smb/client/dir.c index cb10088197d2..6d2378eeb7f6 100644 --- a/fs/smb/client/dir.c +++ b/fs/smb/client/dir.c @@ -82,10 +82,11 @@ char *__build_path_from_dentry_optional_prefix(struct dentry *direntry, void *pa const char *tree, int tree_len, bool prefix) { - int dfsplen; - int pplen = 0; - struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(direntry); + unsigned int sbflags = cifs_sb_flags(cifs_sb); char dirsep = CIFS_DIR_SEP(cifs_sb); + int pplen = 0; + int dfsplen; char *s; if (unlikely(!page)) @@ -96,7 +97,7 @@ char *__build_path_from_dentry_optional_prefix(struct dentry *direntry, void *pa else dfsplen = 0; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) + if (sbflags & CIFS_MOUNT_USE_PREFIX_PATH) pplen = cifs_sb->prepath ? strlen(cifs_sb->prepath) + 1 : 0; s = dentry_path_raw(direntry, page, PATH_MAX); @@ -123,7 +124,7 @@ char *__build_path_from_dentry_optional_prefix(struct dentry *direntry, void *pa if (dfsplen) { s -= dfsplen; memcpy(s, tree, dfsplen); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) { + if (sbflags & CIFS_MOUNT_POSIX_PATHS) { int i; for (i = 0; i < dfsplen; i++) { if (s[i] == '\\') @@ -152,7 +153,7 @@ char *build_path_from_dentry_optional_prefix(struct dentry *direntry, void *page static int check_name(struct dentry *direntry, struct cifs_tcon *tcon) { - struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(direntry); int i; if (unlikely(tcon->fsAttrInfo.MaxPathNameComponentLength && @@ -160,7 +161,7 @@ check_name(struct dentry *direntry, struct cifs_tcon *tcon) le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength))) return -ENAMETOOLONG; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS)) { + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_POSIX_PATHS)) { for (i = 0; i < direntry->d_name.len; i++) { if (direntry->d_name.name[i] == '\\') { cifs_dbg(FYI, "Invalid file name\n"); @@ -181,11 +182,12 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned int rc = -ENOENT; int create_options = CREATE_NOT_DIR; int desired_access; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct cifs_tcon *tcon = tlink_tcon(tlink); const char *full_path; void *page = alloc_dentry_path(); struct inode *newinode = NULL; + unsigned int sbflags = cifs_sb_flags(cifs_sb); int disposition; struct TCP_Server_Info *server = tcon->ses->server; struct cifs_open_parms oparms; @@ -306,6 +308,7 @@ static int cifs_do_create(struct inode *inode, struct dentry *direntry, unsigned goto out; } + create_options |= cifs_open_create_options(oflags, create_options); /* * if we're not using unix extensions, see if we need to set * ATTR_READONLY on the create call @@ -374,7 +377,7 @@ retry_open: .device = 0, }; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { + if (sbflags & CIFS_MOUNT_SET_UID) { args.uid = current_fsuid(); if (inode->i_mode & S_ISGID) args.gid = inode->i_gid; @@ -411,9 +414,9 @@ cifs_create_get_file_info: if (server->ops->set_lease_key) server->ops->set_lease_key(newinode, fid); if ((*oplock & CIFS_CREATE_ACTION) && S_ISREG(newinode->i_mode)) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) + if (sbflags & CIFS_MOUNT_DYNPERM) newinode->i_mode = mode; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { + if (sbflags & CIFS_MOUNT_SET_UID) { newinode->i_uid = current_fsuid(); if (inode->i_mode & S_ISGID) newinode->i_gid = inode->i_gid; @@ -458,18 +461,20 @@ int cifs_atomic_open(struct inode *inode, struct dentry *direntry, struct file *file, unsigned int oflags, umode_t mode) { - int rc; - unsigned int xid; - struct tcon_link *tlink; - struct cifs_tcon *tcon; + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + struct cifs_open_info_data buf = {}; struct TCP_Server_Info *server; - struct cifs_fid fid = {}; + struct cifsFileInfo *file_info; struct cifs_pending_open open; + struct cifs_fid fid = {}; + struct tcon_link *tlink; + struct cifs_tcon *tcon; + unsigned int sbflags; + unsigned int xid; __u32 oplock; - struct cifsFileInfo *file_info; - struct cifs_open_info_data buf = {}; + int rc; - if (unlikely(cifs_forced_shutdown(CIFS_SB(inode->i_sb)))) + if (unlikely(cifs_forced_shutdown(cifs_sb))) return smb_EIO(smb_eio_trace_forced_shutdown); /* @@ -499,7 +504,7 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, cifs_dbg(FYI, "parent inode = 0x%p name is: %pd and dentry = 0x%p\n", inode, direntry, direntry); - tlink = cifs_sb_tlink(CIFS_SB(inode->i_sb)); + tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) { rc = PTR_ERR(tlink); goto out_free_xid; @@ -536,13 +541,13 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, goto out; } - if (file->f_flags & O_DIRECT && - CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) { - if (CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + sbflags = cifs_sb_flags(cifs_sb); + if ((file->f_flags & O_DIRECT) && (sbflags & CIFS_MOUNT_STRICT_IO)) { + if (sbflags & CIFS_MOUNT_NO_BRL) file->f_op = &cifs_file_direct_nobrl_ops; else file->f_op = &cifs_file_direct_ops; - } + } file_info = cifs_new_fileinfo(&fid, file, tlink, oplock, buf.symlink_target); if (file_info == NULL) { diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 18f31d4eb98d..a69e05f86d7e 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -255,7 +255,7 @@ static void cifs_begin_writeback(struct netfs_io_request *wreq) struct cifs_io_request *req = container_of(wreq, struct cifs_io_request, rreq); int ret; - ret = cifs_get_writable_file(CIFS_I(wreq->inode), FIND_WR_ANY, &req->cfile); + ret = cifs_get_writable_file(CIFS_I(wreq->inode), FIND_ANY, &req->cfile); if (ret) { cifs_dbg(VFS, "No writable handle in writepages ret=%d\n", ret); return; @@ -270,7 +270,7 @@ static void cifs_begin_writeback(struct netfs_io_request *wreq) static int cifs_init_request(struct netfs_io_request *rreq, struct file *file) { struct cifs_io_request *req = container_of(rreq, struct cifs_io_request, rreq); - struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(rreq->inode); struct cifsFileInfo *open_file = NULL; rreq->rsize = cifs_sb->ctx->rsize; @@ -281,7 +281,7 @@ static int cifs_init_request(struct netfs_io_request *rreq, struct file *file) open_file = file->private_data; rreq->netfs_priv = file->private_data; req->cfile = cifsFileInfo_get(open_file); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_RWPIDFORWARD) req->pid = req->cfile->pid; } else if (rreq->origin != NETFS_WRITEBACK) { WARN_ON_ONCE(1); @@ -584,15 +584,8 @@ static int cifs_nt_open(const char *full_path, struct inode *inode, struct cifs_ *********************************************************************/ disposition = cifs_get_disposition(f_flags); - /* BB pass O_SYNC flag through on file attributes .. BB */ - - /* O_SYNC also has bit for O_DSYNC so following check picks up either */ - if (f_flags & O_SYNC) - create_options |= CREATE_WRITE_THROUGH; - - if (f_flags & O_DIRECT) - create_options |= CREATE_NO_BUFFER; + create_options |= cifs_open_create_options(f_flags, create_options); retry_open: oparms = (struct cifs_open_parms) { @@ -711,8 +704,6 @@ struct cifsFileInfo *cifs_new_fileinfo(struct cifs_fid *fid, struct file *file, mutex_init(&cfile->fh_mutex); spin_lock_init(&cfile->file_info_lock); - cifs_sb_active(inode->i_sb); - /* * If the server returned a read oplock and we have mandatory brlocks, * set oplock level to None. @@ -767,7 +758,6 @@ static void cifsFileInfo_put_final(struct cifsFileInfo *cifs_file) struct inode *inode = d_inode(cifs_file->dentry); struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifsLockInfo *li, *tmp; - struct super_block *sb = inode->i_sb; /* * Delete any outstanding lock records. We'll lose them when the file @@ -785,7 +775,6 @@ static void cifsFileInfo_put_final(struct cifsFileInfo *cifs_file) cifs_put_tlink(cifs_file->tlink); dput(cifs_file->dentry); - cifs_sb_deactive(sb); kfree(cifs_file->symlink_target); kfree(cifs_file); } @@ -906,7 +895,7 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, * close because it may cause a error when we open this file * again and get at least level II oplock. */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_STRICT_IO) set_bit(CIFS_INO_INVALID_MAPPING, &cifsi->flags); cifs_set_oplock_level(cifsi, 0); } @@ -955,11 +944,11 @@ void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, int cifs_file_flush(const unsigned int xid, struct inode *inode, struct cifsFileInfo *cfile) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct cifs_tcon *tcon; int rc; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOSSYNC) return 0; if (cfile && (OPEN_FMODE(cfile->f_flags) & FMODE_WRITE)) { @@ -967,7 +956,7 @@ int cifs_file_flush(const unsigned int xid, struct inode *inode, return tcon->ses->server->ops->flush(xid, tcon, &cfile->fid); } - rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &cfile); + rc = cifs_get_writable_file(CIFS_I(inode), FIND_ANY, &cfile); if (!rc) { tcon = tlink_tcon(cfile->tlink); rc = tcon->ses->server->ops->flush(xid, tcon, &cfile->fid); @@ -992,7 +981,7 @@ static int cifs_do_truncate(const unsigned int xid, struct dentry *dentry) return -ERESTARTSYS; mapping_set_error(inode->i_mapping, rc); - cfile = find_writable_file(cinode, FIND_WR_FSUID_ONLY); + cfile = find_writable_file(cinode, FIND_FSUID_ONLY); rc = cifs_file_flush(xid, inode, cfile); if (!rc) { if (cfile) { @@ -1004,7 +993,6 @@ static int cifs_do_truncate(const unsigned int xid, struct dentry *dentry) if (!rc) { netfs_resize_file(&cinode->netfs, 0, true); cifs_setsize(inode, 0); - inode->i_blocks = 0; } } if (cfile) @@ -1015,24 +1003,24 @@ static int cifs_do_truncate(const unsigned int xid, struct dentry *dentry) int cifs_open(struct inode *inode, struct file *file) { + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + struct cifs_open_info_data data = {}; + struct cifsFileInfo *cfile = NULL; + struct TCP_Server_Info *server; + struct cifs_pending_open open; + bool posix_open_ok = false; + struct cifs_fid fid = {}; + struct tcon_link *tlink; + struct cifs_tcon *tcon; + const char *full_path; + unsigned int sbflags; int rc = -EACCES; unsigned int xid; __u32 oplock; - struct cifs_sb_info *cifs_sb; - struct TCP_Server_Info *server; - struct cifs_tcon *tcon; - struct tcon_link *tlink; - struct cifsFileInfo *cfile = NULL; void *page; - const char *full_path; - bool posix_open_ok = false; - struct cifs_fid fid = {}; - struct cifs_pending_open open; - struct cifs_open_info_data data = {}; xid = get_xid(); - cifs_sb = CIFS_SB(inode->i_sb); if (unlikely(cifs_forced_shutdown(cifs_sb))) { free_xid(xid); return smb_EIO(smb_eio_trace_forced_shutdown); @@ -1056,9 +1044,9 @@ int cifs_open(struct inode *inode, struct file *file) cifs_dbg(FYI, "inode = 0x%p file flags are 0x%x for %s\n", inode, file->f_flags, full_path); - if (file->f_flags & O_DIRECT && - cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + sbflags = cifs_sb_flags(cifs_sb); + if ((file->f_flags & O_DIRECT) && (sbflags & CIFS_MOUNT_STRICT_IO)) { + if (sbflags & CIFS_MOUNT_NO_BRL) file->f_op = &cifs_file_direct_nobrl_ops; else file->f_op = &cifs_file_direct_ops; @@ -1072,32 +1060,29 @@ int cifs_open(struct inode *inode, struct file *file) /* Get the cached handle as SMB2 close is deferred */ if (OPEN_FMODE(file->f_flags) & FMODE_WRITE) { - rc = cifs_get_writable_path(tcon, full_path, - FIND_WR_FSUID_ONLY | - FIND_WR_NO_PENDING_DELETE, - &cfile); + rc = __cifs_get_writable_file(CIFS_I(inode), + FIND_FSUID_ONLY | + FIND_NO_PENDING_DELETE | + FIND_OPEN_FLAGS, + file->f_flags, &cfile); } else { - rc = cifs_get_readable_path(tcon, full_path, &cfile); + cfile = __find_readable_file(CIFS_I(inode), + FIND_NO_PENDING_DELETE | + FIND_OPEN_FLAGS, + file->f_flags); + rc = cfile ? 0 : -ENOENT; } if (rc == 0) { - unsigned int oflags = file->f_flags & ~(O_CREAT|O_EXCL|O_TRUNC); - unsigned int cflags = cfile->f_flags & ~(O_CREAT|O_EXCL|O_TRUNC); - - if (cifs_convert_flags(oflags, 0) == cifs_convert_flags(cflags, 0) && - (oflags & (O_SYNC|O_DIRECT)) == (cflags & (O_SYNC|O_DIRECT))) { - file->private_data = cfile; - spin_lock(&CIFS_I(inode)->deferred_lock); - cifs_del_deferred_close(cfile); - spin_unlock(&CIFS_I(inode)->deferred_lock); - goto use_cache; - } - _cifsFileInfo_put(cfile, true, false); - } else { - /* hard link on the defeered close file */ - rc = cifs_get_hardlink_path(tcon, inode, file); - if (rc) - cifs_close_deferred_file(CIFS_I(inode)); - } + file->private_data = cfile; + spin_lock(&CIFS_I(inode)->deferred_lock); + cifs_del_deferred_close(cfile); + spin_unlock(&CIFS_I(inode)->deferred_lock); + goto use_cache; + } + /* hard link on the deferred close file */ + rc = cifs_get_hardlink_path(tcon, inode, file); + if (rc) + cifs_close_deferred_file(CIFS_I(inode)); if (server->oplocks) oplock = REQ_OPLOCK; @@ -1209,7 +1194,7 @@ cifs_relock_file(struct cifsFileInfo *cfile) struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); int rc = 0; #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY - struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cinode); #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ down_read_nested(&cinode->lock_sem, SINGLE_DEPTH_NESTING); @@ -1222,7 +1207,7 @@ cifs_relock_file(struct cifsFileInfo *cfile) #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) + ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0)) rc = cifs_push_posix_locks(cfile); else #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ @@ -1318,13 +1303,8 @@ cifs_reopen_file(struct cifsFileInfo *cfile, bool can_flush) rdwr_for_fscache = 1; desired_access = cifs_convert_flags(cfile->f_flags, rdwr_for_fscache); - - /* O_SYNC also has bit for O_DSYNC so following check picks up either */ - if (cfile->f_flags & O_SYNC) - create_options |= CREATE_WRITE_THROUGH; - - if (cfile->f_flags & O_DIRECT) - create_options |= CREATE_NO_BUFFER; + create_options |= cifs_open_create_options(cfile->f_flags, + create_options); if (server->ops->get_lease_key) server->ops->get_lease_key(inode, &cfile->fid); @@ -2011,7 +1991,7 @@ cifs_push_locks(struct cifsFileInfo *cfile) struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); int rc = 0; #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY - struct cifs_sb_info *cifs_sb = CIFS_SB(cfile->dentry->d_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cinode); #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ /* we are going to update can_cache_brlcks here - need a write access */ @@ -2024,7 +2004,7 @@ cifs_push_locks(struct cifsFileInfo *cfile) #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) + ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0)) rc = cifs_push_posix_locks(cfile); else #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ @@ -2428,11 +2408,11 @@ int cifs_flock(struct file *file, int cmd, struct file_lock *fl) cifs_read_flock(fl, &type, &lock, &unlock, &wait_flag, tcon->ses->server); - cifs_sb = CIFS_FILE_SB(file); + cifs_sb = CIFS_SB(file); if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) + ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0)) posix_lck = true; if (!lock && !unlock) { @@ -2455,14 +2435,14 @@ int cifs_flock(struct file *file, int cmd, struct file_lock *fl) int cifs_lock(struct file *file, int cmd, struct file_lock *flock) { - int rc, xid; + struct cifs_sb_info *cifs_sb = CIFS_SB(file); + struct cifsFileInfo *cfile; int lock = 0, unlock = 0; bool wait_flag = false; bool posix_lck = false; - struct cifs_sb_info *cifs_sb; struct cifs_tcon *tcon; - struct cifsFileInfo *cfile; __u32 type; + int rc, xid; rc = -EACCES; xid = get_xid(); @@ -2477,12 +2457,11 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *flock) cifs_read_flock(flock, &type, &lock, &unlock, &wait_flag, tcon->ses->server); - cifs_sb = CIFS_FILE_SB(file); set_bit(CIFS_INO_CLOSE_ON_LOCK, &CIFS_I(d_inode(cfile->dentry))->flags); if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) + ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0)) posix_lck = true; /* * BB add code here to normalize offset and length to account for @@ -2529,14 +2508,37 @@ void cifs_write_subrequest_terminated(struct cifs_io_subrequest *wdata, ssize_t netfs_write_subrequest_terminated(&wdata->subreq, result); } -struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, - bool fsuid_only) +static bool open_flags_match(struct cifsInodeInfo *cinode, + unsigned int oflags, unsigned int cflags) +{ + struct inode *inode = &cinode->netfs.inode; + int crw = 0, orw = 0; + + oflags &= ~(O_CREAT | O_EXCL | O_TRUNC); + cflags &= ~(O_CREAT | O_EXCL | O_TRUNC); + + if (cifs_fscache_enabled(inode)) { + if (OPEN_FMODE(cflags) & FMODE_WRITE) + crw = 1; + if (OPEN_FMODE(oflags) & FMODE_WRITE) + orw = 1; + } + if (cifs_convert_flags(oflags, orw) != cifs_convert_flags(cflags, crw)) + return false; + + return (oflags & (O_SYNC | O_DIRECT)) == (cflags & (O_SYNC | O_DIRECT)); +} + +struct cifsFileInfo *__find_readable_file(struct cifsInodeInfo *cifs_inode, + unsigned int find_flags, + unsigned int open_flags) { + struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode); + bool fsuid_only = find_flags & FIND_FSUID_ONLY; struct cifsFileInfo *open_file = NULL; - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); /* only filter by fsuid on multiuser mounts */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_MULTIUSER)) fsuid_only = false; spin_lock(&cifs_inode->open_file_lock); @@ -2546,6 +2548,13 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, list_for_each_entry(open_file, &cifs_inode->openFileList, flist) { if (fsuid_only && !uid_eq(open_file->uid, current_fsuid())) continue; + if ((find_flags & FIND_NO_PENDING_DELETE) && + open_file->status_file_deleted) + continue; + if ((find_flags & FIND_OPEN_FLAGS) && + !open_flags_match(cifs_inode, open_flags, + open_file->f_flags)) + continue; if (OPEN_FMODE(open_file->f_flags) & FMODE_READ) { if ((!open_file->invalidHandle)) { /* found a good file */ @@ -2564,17 +2573,17 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, } /* Return -EBADF if no handle is found and general rc otherwise */ -int -cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, - struct cifsFileInfo **ret_file) +int __cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, + unsigned int find_flags, unsigned int open_flags, + struct cifsFileInfo **ret_file) { struct cifsFileInfo *open_file, *inv_file = NULL; struct cifs_sb_info *cifs_sb; bool any_available = false; int rc = -EBADF; unsigned int refind = 0; - bool fsuid_only = flags & FIND_WR_FSUID_ONLY; - bool with_delete = flags & FIND_WR_WITH_DELETE; + bool fsuid_only = find_flags & FIND_FSUID_ONLY; + bool with_delete = find_flags & FIND_WITH_DELETE; *ret_file = NULL; /* @@ -2589,10 +2598,10 @@ cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, return rc; } - cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); + cifs_sb = CIFS_SB(cifs_inode); /* only filter by fsuid on multiuser mounts */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_MULTIUSER)) fsuid_only = false; spin_lock(&cifs_inode->open_file_lock); @@ -2608,9 +2617,13 @@ refind_writable: continue; if (with_delete && !(open_file->fid.access & DELETE)) continue; - if ((flags & FIND_WR_NO_PENDING_DELETE) && + if ((find_flags & FIND_NO_PENDING_DELETE) && open_file->status_file_deleted) continue; + if ((find_flags & FIND_OPEN_FLAGS) && + !open_flags_match(cifs_inode, open_flags, + open_file->f_flags)) + continue; if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) { if (!open_file->invalidHandle) { /* found a good writable file */ @@ -2727,17 +2740,7 @@ cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, cinode = CIFS_I(d_inode(cfile->dentry)); spin_unlock(&tcon->open_file_lock); free_dentry_path(page); - *ret_file = find_readable_file(cinode, 0); - if (*ret_file) { - spin_lock(&cinode->open_file_lock); - if ((*ret_file)->status_file_deleted) { - spin_unlock(&cinode->open_file_lock); - cifsFileInfo_put(*ret_file); - *ret_file = NULL; - } else { - spin_unlock(&cinode->open_file_lock); - } - } + *ret_file = find_readable_file(cinode, FIND_ANY); return *ret_file ? 0 : -ENOENT; } @@ -2787,7 +2790,7 @@ int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync) struct TCP_Server_Info *server; struct cifsFileInfo *smbfile = file->private_data; struct inode *inode = file_inode(file); - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file); + struct cifs_sb_info *cifs_sb = CIFS_SB(file); rc = file_write_and_wait_range(file, start, end); if (rc) { @@ -2801,7 +2804,7 @@ int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync) file, datasync); tcon = tlink_tcon(smbfile->tlink); - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) { + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOSSYNC)) { server = tcon->ses->server; if (server->ops->flush == NULL) { rc = -ENOSYS; @@ -2809,7 +2812,7 @@ int cifs_fsync(struct file *file, loff_t start, loff_t end, int datasync) } if ((OPEN_FMODE(smbfile->f_flags) & FMODE_WRITE) == 0) { - smbfile = find_writable_file(CIFS_I(inode), FIND_WR_ANY); + smbfile = find_writable_file(CIFS_I(inode), FIND_ANY); if (smbfile) { rc = server->ops->flush(xid, tcon, &smbfile->fid); cifsFileInfo_put(smbfile); @@ -2853,7 +2856,7 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) struct inode *inode = file->f_mapping->host; struct cifsInodeInfo *cinode = CIFS_I(inode); struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); ssize_t rc; rc = netfs_start_io_write(inode); @@ -2870,7 +2873,7 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) if (rc <= 0) goto out; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) && + if ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) && (cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(from), server->vals->exclusive_lock_type, 0, NULL, CIFS_WRITE_OP))) { @@ -2893,7 +2896,7 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from) { struct inode *inode = file_inode(iocb->ki_filp); struct cifsInodeInfo *cinode = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct cifsFileInfo *cfile = (struct cifsFileInfo *) iocb->ki_filp->private_data; struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); @@ -2906,7 +2909,7 @@ cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from) if (CIFS_CACHE_WRITE(cinode)) { if (cap_unix(tcon->ses) && (CIFS_UNIX_FCNTL_CAP & le64_to_cpu(tcon->fsUnixInfo.Capability)) && - ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0)) { + ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0)) { written = netfs_file_write_iter(iocb, from); goto out; } @@ -2994,7 +2997,7 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) { struct inode *inode = file_inode(iocb->ki_filp); struct cifsInodeInfo *cinode = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct cifsFileInfo *cfile = (struct cifsFileInfo *) iocb->ki_filp->private_data; struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); @@ -3011,7 +3014,7 @@ cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) if (!CIFS_CACHE_READ(cinode)) return netfs_unbuffered_read_iter(iocb, to); - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOPOSIXBRL) == 0) { + if ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NOPOSIXBRL) == 0) { if (iocb->ki_flags & IOCB_DIRECT) return netfs_unbuffered_read_iter(iocb, to); return netfs_buffered_read_iter(iocb, to); @@ -3130,10 +3133,9 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file, if (is_inode_writable(cifsInode) || ((cifsInode->oplock & CIFS_CACHE_RW_FLG) != 0 && from_readdir)) { /* This inode is open for write at least once */ - struct cifs_sb_info *cifs_sb; + struct cifs_sb_info *cifs_sb = CIFS_SB(cifsInode); - cifs_sb = CIFS_SB(cifsInode->netfs.inode.i_sb); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_DIRECT_IO) { /* since no page cache to corrupt on directio we can change size safely */ return true; @@ -3165,12 +3167,6 @@ void cifs_oplock_break(struct work_struct *work) __u64 persistent_fid, volatile_fid; __u16 net_fid; - /* - * Hold a reference to the superblock to prevent it and its inodes from - * being freed while we are accessing cinode. Otherwise, _cifsFileInfo_put() - * may release the last reference to the sb and trigger inode eviction. - */ - cifs_sb_active(sb); wait_on_bit(&cinode->flags, CIFS_INODE_PENDING_WRITERS, TASK_UNINTERRUPTIBLE); @@ -3181,7 +3177,7 @@ void cifs_oplock_break(struct work_struct *work) server = tcon->ses->server; scoped_guard(spinlock, &cinode->open_file_lock) { - unsigned int sbflags = cifs_sb->mnt_cifs_flags; + unsigned int sbflags = cifs_sb_flags(cifs_sb); server->ops->downgrade_oplock(server, cinode, cfile->oplock_level, cfile->oplock_epoch, &purge_cache); @@ -3255,7 +3251,6 @@ oplock_break_ack: cifs_put_tlink(tlink); out: cifs_done_oplock_break(cinode); - cifs_sb_deactive(sb); } static int cifs_swap_activate(struct swap_info_struct *sis, diff --git a/fs/smb/client/fs_context.c b/fs/smb/client/fs_context.c index 09fe749e97ee..a4a7c7eee038 100644 --- a/fs/smb/client/fs_context.c +++ b/fs/smb/client/fs_context.c @@ -1997,7 +1997,7 @@ int smb3_init_fs_context(struct fs_context *fc) ctx->backupuid_specified = false; /* no backup intent for a user */ ctx->backupgid_specified = false; /* no backup intent for a group */ - ctx->retrans = 1; + ctx->retrans = 0; ctx->reparse_type = CIFS_REPARSE_TYPE_DEFAULT; ctx->symlink_type = CIFS_SYMLINK_TYPE_DEFAULT; ctx->nonativesocket = 0; @@ -2062,161 +2062,160 @@ smb3_cleanup_fs_context(struct smb3_fs_context *ctx) kfree(ctx); } -void smb3_update_mnt_flags(struct cifs_sb_info *cifs_sb) +unsigned int smb3_update_mnt_flags(struct cifs_sb_info *cifs_sb) { + unsigned int sbflags = cifs_sb_flags(cifs_sb); struct smb3_fs_context *ctx = cifs_sb->ctx; if (ctx->nodfs) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_DFS; + sbflags |= CIFS_MOUNT_NO_DFS; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NO_DFS; + sbflags &= ~CIFS_MOUNT_NO_DFS; if (ctx->noperm) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_PERM; + sbflags |= CIFS_MOUNT_NO_PERM; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NO_PERM; + sbflags &= ~CIFS_MOUNT_NO_PERM; if (ctx->setuids) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_SET_UID; + sbflags |= CIFS_MOUNT_SET_UID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SET_UID; + sbflags &= ~CIFS_MOUNT_SET_UID; if (ctx->setuidfromacl) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_UID_FROM_ACL; + sbflags |= CIFS_MOUNT_UID_FROM_ACL; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_UID_FROM_ACL; + sbflags &= ~CIFS_MOUNT_UID_FROM_ACL; if (ctx->server_ino) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_SERVER_INUM; + sbflags |= CIFS_MOUNT_SERVER_INUM; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM; + sbflags &= ~CIFS_MOUNT_SERVER_INUM; if (ctx->remap) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MAP_SFM_CHR; + sbflags |= CIFS_MOUNT_MAP_SFM_CHR; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_MAP_SFM_CHR; + sbflags &= ~CIFS_MOUNT_MAP_SFM_CHR; if (ctx->sfu_remap) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MAP_SPECIAL_CHR; + sbflags |= CIFS_MOUNT_MAP_SPECIAL_CHR; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_MAP_SPECIAL_CHR; + sbflags &= ~CIFS_MOUNT_MAP_SPECIAL_CHR; if (ctx->no_xattr) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_XATTR; + sbflags |= CIFS_MOUNT_NO_XATTR; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NO_XATTR; + sbflags &= ~CIFS_MOUNT_NO_XATTR; if (ctx->sfu_emul) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_UNX_EMUL; + sbflags |= CIFS_MOUNT_UNX_EMUL; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_UNX_EMUL; + sbflags &= ~CIFS_MOUNT_UNX_EMUL; if (ctx->nobrl) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_BRL; + sbflags |= CIFS_MOUNT_NO_BRL; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NO_BRL; + sbflags &= ~CIFS_MOUNT_NO_BRL; if (ctx->nohandlecache) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_HANDLE_CACHE; + sbflags |= CIFS_MOUNT_NO_HANDLE_CACHE; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NO_HANDLE_CACHE; + sbflags &= ~CIFS_MOUNT_NO_HANDLE_CACHE; if (ctx->nostrictsync) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOSSYNC; + sbflags |= CIFS_MOUNT_NOSSYNC; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NOSSYNC; + sbflags &= ~CIFS_MOUNT_NOSSYNC; if (ctx->mand_lock) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NOPOSIXBRL; + sbflags |= CIFS_MOUNT_NOPOSIXBRL; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_NOPOSIXBRL; + sbflags &= ~CIFS_MOUNT_NOPOSIXBRL; if (ctx->rwpidforward) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_RWPIDFORWARD; + sbflags |= CIFS_MOUNT_RWPIDFORWARD; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_RWPIDFORWARD; + sbflags &= ~CIFS_MOUNT_RWPIDFORWARD; if (ctx->mode_ace) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MODE_FROM_SID; + sbflags |= CIFS_MOUNT_MODE_FROM_SID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_MODE_FROM_SID; + sbflags &= ~CIFS_MOUNT_MODE_FROM_SID; if (ctx->cifs_acl) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_ACL; + sbflags |= CIFS_MOUNT_CIFS_ACL; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_CIFS_ACL; + sbflags &= ~CIFS_MOUNT_CIFS_ACL; if (ctx->backupuid_specified) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_BACKUPUID; + sbflags |= CIFS_MOUNT_CIFS_BACKUPUID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_CIFS_BACKUPUID; + sbflags &= ~CIFS_MOUNT_CIFS_BACKUPUID; if (ctx->backupgid_specified) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_CIFS_BACKUPGID; + sbflags |= CIFS_MOUNT_CIFS_BACKUPGID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_CIFS_BACKUPGID; + sbflags &= ~CIFS_MOUNT_CIFS_BACKUPGID; if (ctx->override_uid) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_OVERR_UID; + sbflags |= CIFS_MOUNT_OVERR_UID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_OVERR_UID; + sbflags &= ~CIFS_MOUNT_OVERR_UID; if (ctx->override_gid) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_OVERR_GID; + sbflags |= CIFS_MOUNT_OVERR_GID; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_OVERR_GID; + sbflags &= ~CIFS_MOUNT_OVERR_GID; if (ctx->dynperm) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DYNPERM; + sbflags |= CIFS_MOUNT_DYNPERM; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_DYNPERM; + sbflags &= ~CIFS_MOUNT_DYNPERM; if (ctx->fsc) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_FSCACHE; + sbflags |= CIFS_MOUNT_FSCACHE; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_FSCACHE; + sbflags &= ~CIFS_MOUNT_FSCACHE; if (ctx->multiuser) - cifs_sb->mnt_cifs_flags |= (CIFS_MOUNT_MULTIUSER | - CIFS_MOUNT_NO_PERM); + sbflags |= CIFS_MOUNT_MULTIUSER | CIFS_MOUNT_NO_PERM; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_MULTIUSER; + sbflags &= ~CIFS_MOUNT_MULTIUSER; if (ctx->strict_io) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_STRICT_IO; + sbflags |= CIFS_MOUNT_STRICT_IO; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_STRICT_IO; + sbflags &= ~CIFS_MOUNT_STRICT_IO; if (ctx->direct_io) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DIRECT_IO; + sbflags |= CIFS_MOUNT_DIRECT_IO; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_DIRECT_IO; + sbflags &= ~CIFS_MOUNT_DIRECT_IO; if (ctx->mfsymlinks) - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MF_SYMLINKS; + sbflags |= CIFS_MOUNT_MF_SYMLINKS; else - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_MF_SYMLINKS; - if (ctx->mfsymlinks) { - if (ctx->sfu_emul) { - /* - * Our SFU ("Services for Unix") emulation allows now - * creating new and reading existing SFU symlinks. - * Older Linux kernel versions were not able to neither - * read existing nor create new SFU symlinks. But - * creating and reading SFU style mknod and FIFOs was - * supported for long time. When "mfsymlinks" and - * "sfu" are both enabled at the same time, it allows - * reading both types of symlinks, but will only create - * them with mfsymlinks format. This allows better - * Apple compatibility, compatibility with older Linux - * kernel clients (probably better for Samba too) - * while still recognizing old Windows style symlinks. - */ - cifs_dbg(VFS, "mount options mfsymlinks and sfu both enabled\n"); - } - } - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SHUTDOWN; + sbflags &= ~CIFS_MOUNT_MF_SYMLINKS; - return; + if (ctx->mfsymlinks && ctx->sfu_emul) { + /* + * Our SFU ("Services for Unix") emulation allows now + * creating new and reading existing SFU symlinks. + * Older Linux kernel versions were not able to neither + * read existing nor create new SFU symlinks. But + * creating and reading SFU style mknod and FIFOs was + * supported for long time. When "mfsymlinks" and + * "sfu" are both enabled at the same time, it allows + * reading both types of symlinks, but will only create + * them with mfsymlinks format. This allows better + * Apple compatibility, compatibility with older Linux + * kernel clients (probably better for Samba too) + * while still recognizing old Windows style symlinks. + */ + cifs_dbg(VFS, "mount options mfsymlinks and sfu both enabled\n"); + } + sbflags &= ~CIFS_MOUNT_SHUTDOWN; + atomic_set(&cifs_sb->mnt_cifs_flags, sbflags); + return sbflags; } diff --git a/fs/smb/client/fs_context.h b/fs/smb/client/fs_context.h index 49b2a6f09ca2..0b64fcb5d302 100644 --- a/fs/smb/client/fs_context.h +++ b/fs/smb/client/fs_context.h @@ -374,7 +374,7 @@ int smb3_fs_context_dup(struct smb3_fs_context *new_ctx, struct smb3_fs_context *ctx); int smb3_sync_session_ctx_passwords(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses); -void smb3_update_mnt_flags(struct cifs_sb_info *cifs_sb); +unsigned int smb3_update_mnt_flags(struct cifs_sb_info *cifs_sb); /* * max deferred close timeout (jiffies) - 2^30 diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index d4d3cfeb6c90..888f9e35f14b 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -40,32 +40,33 @@ static void cifs_set_netfs_context(struct inode *inode) static void cifs_set_ops(struct inode *inode) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct netfs_inode *ictx = netfs_inode(inode); + unsigned int sbflags = cifs_sb_flags(cifs_sb); switch (inode->i_mode & S_IFMT) { case S_IFREG: inode->i_op = &cifs_file_inode_ops; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if (sbflags & CIFS_MOUNT_DIRECT_IO) { set_bit(NETFS_ICTX_UNBUFFERED, &ictx->flags); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + if (sbflags & CIFS_MOUNT_NO_BRL) inode->i_fop = &cifs_file_direct_nobrl_ops; else inode->i_fop = &cifs_file_direct_ops; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_STRICT_IO) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + } else if (sbflags & CIFS_MOUNT_STRICT_IO) { + if (sbflags & CIFS_MOUNT_NO_BRL) inode->i_fop = &cifs_file_strict_nobrl_ops; else inode->i_fop = &cifs_file_strict_ops; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + } else if (sbflags & CIFS_MOUNT_NO_BRL) inode->i_fop = &cifs_file_nobrl_ops; else { /* not direct, send byte range locks */ inode->i_fop = &cifs_file_ops; } /* check if server can support readahead */ - if (cifs_sb_master_tcon(cifs_sb)->ses->server->max_read < - PAGE_SIZE + MAX_CIFS_HDR_SIZE) + if (tcon->ses->server->max_read < PAGE_SIZE + MAX_CIFS_HDR_SIZE) inode->i_data.a_ops = &cifs_addr_ops_smallbuf; else inode->i_data.a_ops = &cifs_addr_ops; @@ -194,8 +195,8 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr, inode->i_gid = fattr->cf_gid; /* if dynperm is set, don't clobber existing mode */ - if (inode_state_read(inode) & I_NEW || - !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)) + if ((inode_state_read(inode) & I_NEW) || + !(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_DYNPERM)) inode->i_mode = fattr->cf_mode; cifs_i->cifsAttrs = fattr->cf_cifsattrs; @@ -218,13 +219,7 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr, */ if (is_size_safe_to_change(cifs_i, fattr->cf_eof, from_readdir)) { i_size_write(inode, fattr->cf_eof); - - /* - * i_blocks is not related to (i_size / i_blksize), - * but instead 512 byte (2**9) size is required for - * calculating num blocks. - */ - inode->i_blocks = (512 - 1 + fattr->cf_bytes) >> 9; + inode->i_blocks = CIFS_INO_BLOCKS(fattr->cf_bytes); } if (S_ISLNK(fattr->cf_mode) && fattr->cf_symlink_target) { @@ -248,10 +243,8 @@ cifs_fill_uniqueid(struct super_block *sb, struct cifs_fattr *fattr) { struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) - return; - - fattr->cf_uniqueid = iunique(sb, ROOT_I); + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM)) + fattr->cf_uniqueid = iunique(sb, ROOT_I); } /* Fill a cifs_fattr struct with info from FILE_UNIX_BASIC_INFO. */ @@ -259,6 +252,8 @@ void cifs_unix_basic_to_fattr(struct cifs_fattr *fattr, FILE_UNIX_BASIC_INFO *info, struct cifs_sb_info *cifs_sb) { + unsigned int sbflags; + memset(fattr, 0, sizeof(*fattr)); fattr->cf_uniqueid = le64_to_cpu(info->UniqueId); fattr->cf_bytes = le64_to_cpu(info->NumOfBytes); @@ -317,8 +312,9 @@ cifs_unix_basic_to_fattr(struct cifs_fattr *fattr, FILE_UNIX_BASIC_INFO *info, break; } + sbflags = cifs_sb_flags(cifs_sb); fattr->cf_uid = cifs_sb->ctx->linux_uid; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)) { + if (!(sbflags & CIFS_MOUNT_OVERR_UID)) { u64 id = le64_to_cpu(info->Uid); if (id < ((uid_t)-1)) { kuid_t uid = make_kuid(&init_user_ns, id); @@ -328,7 +324,7 @@ cifs_unix_basic_to_fattr(struct cifs_fattr *fattr, FILE_UNIX_BASIC_INFO *info, } fattr->cf_gid = cifs_sb->ctx->linux_gid; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)) { + if (!(sbflags & CIFS_MOUNT_OVERR_GID)) { u64 id = le64_to_cpu(info->Gid); if (id < ((gid_t)-1)) { kgid_t gid = make_kgid(&init_user_ns, id); @@ -382,7 +378,7 @@ static int update_inode_info(struct super_block *sb, * * If file type or uniqueid is different, return error. */ - if (unlikely((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) && + if (unlikely((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM) && CIFS_I(*inode)->uniqueid != fattr->cf_uniqueid)) { CIFS_I(*inode)->time = 0; /* force reval */ return -ESTALE; @@ -468,7 +464,7 @@ static int cifs_get_unix_fattr(const unsigned char *full_path, cifs_fill_uniqueid(sb, fattr); /* check for Minshall+French symlinks */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) { + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_MF_SYMLINKS) { tmprc = check_mf_symlink(xid, tcon, cifs_sb, fattr, full_path); cifs_dbg(FYI, "check_mf_symlink: %d\n", tmprc); } @@ -1081,7 +1077,7 @@ cifs_backup_query_path_info(int xid, else if ((tcon->ses->capabilities & tcon->ses->server->vals->cap_nt_find) == 0) info.info_level = SMB_FIND_FILE_INFO_STANDARD; - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) + else if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM) info.info_level = SMB_FIND_FILE_ID_FULL_DIR_INFO; else /* no srvino useful for fallback to some netapp */ info.info_level = SMB_FIND_FILE_DIRECTORY_INFO; @@ -1109,7 +1105,7 @@ static void cifs_set_fattr_ino(int xid, struct cifs_tcon *tcon, struct super_blo struct TCP_Server_Info *server = tcon->ses->server; int rc; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) { + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM)) { if (*inode) fattr->cf_uniqueid = CIFS_I(*inode)->uniqueid; else @@ -1263,14 +1259,15 @@ static int cifs_get_fattr(struct cifs_open_info_data *data, struct inode **inode, const char *full_path) { + struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct cifs_open_info_data tmp_data = {}; - struct cifs_tcon *tcon; + void *smb1_backup_rsp_buf = NULL; struct TCP_Server_Info *server; + struct cifs_tcon *tcon; struct tcon_link *tlink; - struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - void *smb1_backup_rsp_buf = NULL; - int rc = 0; + unsigned int sbflags; int tmprc = 0; + int rc = 0; tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) @@ -1370,16 +1367,17 @@ static int cifs_get_fattr(struct cifs_open_info_data *data, #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY handle_mnt_opt: #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ + sbflags = cifs_sb_flags(cifs_sb); /* query for SFU type info if supported and needed */ if ((fattr->cf_cifsattrs & ATTR_SYSTEM) && - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)) { + (sbflags & CIFS_MOUNT_UNX_EMUL)) { tmprc = cifs_sfu_type(fattr, full_path, cifs_sb, xid); if (tmprc) cifs_dbg(FYI, "cifs_sfu_type failed: %d\n", tmprc); } /* fill in 0777 bits from ACL */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) { + if (sbflags & CIFS_MOUNT_MODE_FROM_SID) { rc = cifs_acl_to_fattr(cifs_sb, fattr, *inode, true, full_path, fid); if (rc == -EREMOTE) @@ -1389,7 +1387,7 @@ handle_mnt_opt: __func__, rc); goto out; } - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { + } else if (sbflags & CIFS_MOUNT_CIFS_ACL) { rc = cifs_acl_to_fattr(cifs_sb, fattr, *inode, false, full_path, fid); if (rc == -EREMOTE) @@ -1399,7 +1397,7 @@ handle_mnt_opt: __func__, rc); goto out; } - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) + } else if (sbflags & CIFS_MOUNT_UNX_EMUL) /* fill in remaining high mode bits e.g. SUID, VTX */ cifs_sfu_mode(fattr, full_path, cifs_sb, xid); else if (!(tcon->posix_extensions)) @@ -1409,7 +1407,7 @@ handle_mnt_opt: /* check for Minshall+French symlinks */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) { + if (sbflags & CIFS_MOUNT_MF_SYMLINKS) { tmprc = check_mf_symlink(xid, tcon, cifs_sb, fattr, full_path); cifs_dbg(FYI, "check_mf_symlink: %d\n", tmprc); } @@ -1509,7 +1507,7 @@ static int smb311_posix_get_fattr(struct cifs_open_info_data *data, * 3. Tweak fattr based on mount options */ /* check for Minshall+French symlinks */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) { + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_MF_SYMLINKS) { tmprc = check_mf_symlink(xid, tcon, cifs_sb, fattr, full_path); cifs_dbg(FYI, "check_mf_symlink: %d\n", tmprc); } @@ -1660,7 +1658,7 @@ struct inode *cifs_root_iget(struct super_block *sb) int len; int rc; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH) + if ((cifs_sb_flags(cifs_sb) & CIFS_MOUNT_USE_PREFIX_PATH) && cifs_sb->prepath) { len = strlen(cifs_sb->prepath); path = kzalloc(len + 2 /* leading sep + null */, GFP_KERNEL); @@ -2098,8 +2096,9 @@ cifs_mkdir_qinfo(struct inode *parent, struct dentry *dentry, umode_t mode, const char *full_path, struct cifs_sb_info *cifs_sb, struct cifs_tcon *tcon, const unsigned int xid) { - int rc = 0; struct inode *inode = NULL; + unsigned int sbflags; + int rc = 0; if (tcon->posix_extensions) { rc = smb311_posix_get_inode_info(&inode, full_path, @@ -2139,6 +2138,7 @@ cifs_mkdir_qinfo(struct inode *parent, struct dentry *dentry, umode_t mode, if (parent->i_mode & S_ISGID) mode |= S_ISGID; + sbflags = cifs_sb_flags(cifs_sb); #ifdef CONFIG_CIFS_ALLOW_INSECURE_LEGACY if (tcon->unix_ext) { struct cifs_unix_set_info_args args = { @@ -2148,7 +2148,7 @@ cifs_mkdir_qinfo(struct inode *parent, struct dentry *dentry, umode_t mode, .mtime = NO_CHANGE_64, .device = 0, }; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { + if (sbflags & CIFS_MOUNT_SET_UID) { args.uid = current_fsuid(); if (parent->i_mode & S_ISGID) args.gid = parent->i_gid; @@ -2166,14 +2166,14 @@ cifs_mkdir_qinfo(struct inode *parent, struct dentry *dentry, umode_t mode, { #endif /* CONFIG_CIFS_ALLOW_INSECURE_LEGACY */ struct TCP_Server_Info *server = tcon->ses->server; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) && + if (!(sbflags & CIFS_MOUNT_CIFS_ACL) && (mode & S_IWUGO) == 0 && server->ops->mkdir_setinfo) server->ops->mkdir_setinfo(inode, full_path, cifs_sb, tcon, xid); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) + if (sbflags & CIFS_MOUNT_DYNPERM) inode->i_mode = (mode | S_IFDIR); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { + if (sbflags & CIFS_MOUNT_SET_UID) { inode->i_uid = current_fsuid(); if (inode->i_mode & S_ISGID) inode->i_gid = parent->i_gid; @@ -2686,7 +2686,7 @@ cifs_dentry_needs_reval(struct dentry *dentry) { struct inode *inode = d_inode(dentry); struct cifsInodeInfo *cifs_i = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct cached_fid *cfid = NULL; @@ -2727,7 +2727,7 @@ cifs_dentry_needs_reval(struct dentry *dentry) } /* hardlinked files w/ noserverino get "special" treatment */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) && + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM) && S_ISREG(inode->i_mode) && inode->i_nlink != 1) return true; @@ -2752,10 +2752,10 @@ cifs_wait_bit_killable(struct wait_bit_key *key, int mode) int cifs_revalidate_mapping(struct inode *inode) { - int rc; struct cifsInodeInfo *cifs_inode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); unsigned long *flags = &cifs_inode->flags; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + int rc; /* swapfiles are not supposed to be shared */ if (IS_SWAPFILE(inode)) @@ -2768,7 +2768,7 @@ cifs_revalidate_mapping(struct inode *inode) if (test_and_clear_bit(CIFS_INO_INVALID_MAPPING, flags)) { /* for cache=singleclient, do not invalidate */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_RW_CACHE) goto skip_invalidate; cifs_inode->netfs.zero_point = cifs_inode->netfs.remote_i_size; @@ -2892,10 +2892,11 @@ int cifs_revalidate_dentry(struct dentry *dentry) int cifs_getattr(struct mnt_idmap *idmap, const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { - struct dentry *dentry = path->dentry; - struct cifs_sb_info *cifs_sb = CIFS_SB(dentry->d_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(path->dentry); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + struct dentry *dentry = path->dentry; struct inode *inode = d_inode(dentry); + unsigned int sbflags; int rc; if (unlikely(cifs_forced_shutdown(CIFS_SB(inode->i_sb)))) @@ -2952,12 +2953,13 @@ int cifs_getattr(struct mnt_idmap *idmap, const struct path *path, * enabled, and the admin hasn't overridden them, set the ownership * to the fsuid/fsgid of the current process. */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) && - !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) && + sbflags = cifs_sb_flags(cifs_sb); + if ((sbflags & CIFS_MOUNT_MULTIUSER) && + !(sbflags & CIFS_MOUNT_CIFS_ACL) && !tcon->unix_ext) { - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)) + if (!(sbflags & CIFS_MOUNT_OVERR_UID)) stat->uid = current_fsuid(); - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)) + if (!(sbflags & CIFS_MOUNT_OVERR_GID)) stat->gid = current_fsgid(); } return 0; @@ -2989,7 +2991,7 @@ int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start, } } - cfile = find_readable_file(cifs_i, false); + cfile = find_readable_file(cifs_i, FIND_ANY); if (cfile == NULL) return -EINVAL; @@ -3007,6 +3009,11 @@ void cifs_setsize(struct inode *inode, loff_t offset) { spin_lock(&inode->i_lock); i_size_write(inode, offset); + /* + * Until we can query the server for actual allocation size, + * this is best estimate we have for blocks allocated for a file. + */ + inode->i_blocks = CIFS_INO_BLOCKS(offset); spin_unlock(&inode->i_lock); inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); truncate_pagecache(inode, offset); @@ -3042,7 +3049,7 @@ int cifs_file_set_size(const unsigned int xid, struct dentry *dentry, size, false); cifs_dbg(FYI, "%s: set_file_size: rc = %d\n", __func__, rc); } else { - open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY); + open_file = find_writable_file(cifsInode, FIND_FSUID_ONLY); if (open_file) { tcon = tlink_tcon(open_file->tlink); server = tcon->ses->server; @@ -3079,14 +3086,6 @@ set_size_out: if (rc == 0) { netfs_resize_file(&cifsInode->netfs, size, true); cifs_setsize(inode, size); - /* - * i_blocks is not related to (i_size / i_blksize), but instead - * 512 byte (2**9) size is required for calculating num blocks. - * Until we can query the server for actual allocation size, - * this is best estimate we have for blocks allocated for a file - * Number of blocks must be rounded up so size 1 is not 0 blocks - */ - inode->i_blocks = (512 - 1 + size) >> 9; } return rc; @@ -3102,7 +3101,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) void *page = alloc_dentry_path(); struct inode *inode = d_inode(direntry); struct cifsInodeInfo *cifsInode = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); struct tcon_link *tlink; struct cifs_tcon *pTcon; struct cifs_unix_set_info_args *args = NULL; @@ -3113,7 +3112,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) xid = get_xid(); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_PERM) attrs->ia_valid |= ATTR_FORCE; rc = setattr_prepare(&nop_mnt_idmap, direntry, attrs); @@ -3211,7 +3210,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) open_file->fid.netfid, open_file->pid); } else { - open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY); + open_file = find_writable_file(cifsInode, FIND_FSUID_ONLY); if (open_file) { pTcon = tlink_tcon(open_file->tlink); rc = CIFSSMBUnixSetFileInfo(xid, pTcon, args, @@ -3266,26 +3265,26 @@ out: static int cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) { - unsigned int xid; - kuid_t uid = INVALID_UID; - kgid_t gid = INVALID_GID; struct inode *inode = d_inode(direntry); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifsInodeInfo *cifsInode = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + unsigned int sbflags = cifs_sb_flags(cifs_sb); struct cifsFileInfo *cfile = NULL; - const char *full_path; void *page = alloc_dentry_path(); - int rc = -EACCES; - __u32 dosattr = 0; __u64 mode = NO_CHANGE_64; - bool posix = cifs_sb_master_tcon(cifs_sb)->posix_extensions; + kuid_t uid = INVALID_UID; + kgid_t gid = INVALID_GID; + const char *full_path; + __u32 dosattr = 0; + int rc = -EACCES; + unsigned int xid; xid = get_xid(); cifs_dbg(FYI, "setattr on file %pd attrs->ia_valid 0x%x\n", direntry, attrs->ia_valid); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) + if (sbflags & CIFS_MOUNT_NO_PERM) attrs->ia_valid |= ATTR_FORCE; rc = setattr_prepare(&nop_mnt_idmap, direntry, attrs); @@ -3346,8 +3345,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (attrs->ia_valid & ATTR_GID) gid = attrs->ia_gid; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) || - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID)) { + if (sbflags & (CIFS_MOUNT_CIFS_ACL | CIFS_MOUNT_MODE_FROM_SID)) { if (uid_valid(uid) || gid_valid(gid)) { mode = NO_CHANGE_64; rc = id_mode_to_cifs_acl(inode, full_path, &mode, @@ -3358,9 +3356,9 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) goto cifs_setattr_exit; } } - } else - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)) + } else if (!(sbflags & CIFS_MOUNT_SET_UID)) { attrs->ia_valid &= ~(ATTR_UID | ATTR_GID); + } /* skip mode change if it's just for clearing setuid/setgid */ if (attrs->ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) @@ -3369,9 +3367,8 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (attrs->ia_valid & ATTR_MODE) { mode = attrs->ia_mode; rc = 0; - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) || - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) || - posix) { + if ((sbflags & (CIFS_MOUNT_CIFS_ACL | CIFS_MOUNT_MODE_FROM_SID)) || + cifs_sb_master_tcon(cifs_sb)->posix_extensions) { rc = id_mode_to_cifs_acl(inode, full_path, &mode, INVALID_UID, INVALID_GID); if (rc) { @@ -3393,7 +3390,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) dosattr = cifsInode->cifsAttrs | ATTR_READONLY; /* fix up mode if we're not using dynperm */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM) == 0) + if ((sbflags & CIFS_MOUNT_DYNPERM) == 0) attrs->ia_mode = inode->i_mode & ~S_IWUGO; } else if ((mode & S_IWUGO) && (cifsInode->cifsAttrs & ATTR_READONLY)) { @@ -3404,7 +3401,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) dosattr |= ATTR_NORMAL; /* reset local inode permissions to normal */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)) { + if (!(sbflags & CIFS_MOUNT_DYNPERM)) { attrs->ia_mode &= ~(S_IALLUGO); if (S_ISDIR(inode->i_mode)) attrs->ia_mode |= @@ -3413,7 +3410,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) attrs->ia_mode |= cifs_sb->ctx->file_mode; } - } else if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)) { + } else if (!(sbflags & CIFS_MOUNT_DYNPERM)) { /* ignore mode change - ATTR_READONLY hasn't changed */ attrs->ia_valid &= ~ATTR_MODE; } diff --git a/fs/smb/client/ioctl.c b/fs/smb/client/ioctl.c index 8dc2651e237f..9afab3237e54 100644 --- a/fs/smb/client/ioctl.c +++ b/fs/smb/client/ioctl.c @@ -216,7 +216,7 @@ static int cifs_shutdown(struct super_block *sb, unsigned long arg) */ case CIFS_GOING_FLAGS_LOGFLUSH: case CIFS_GOING_FLAGS_NOLOGFLUSH: - sbi->mnt_cifs_flags |= CIFS_MOUNT_SHUTDOWN; + atomic_or(CIFS_MOUNT_SHUTDOWN, &sbi->mnt_cifs_flags); goto shutdown_good; default: rc = -EINVAL; diff --git a/fs/smb/client/link.c b/fs/smb/client/link.c index a2f7bfa8ad1e..434e8fe74080 100644 --- a/fs/smb/client/link.c +++ b/fs/smb/client/link.c @@ -544,14 +544,15 @@ int cifs_symlink(struct mnt_idmap *idmap, struct inode *inode, struct dentry *direntry, const char *symname) { - int rc = -EOPNOTSUPP; - unsigned int xid; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + struct inode *newinode = NULL; struct tcon_link *tlink; struct cifs_tcon *pTcon; const char *full_path; + int rc = -EOPNOTSUPP; + unsigned int sbflags; + unsigned int xid; void *page; - struct inode *newinode = NULL; if (unlikely(cifs_forced_shutdown(cifs_sb))) return smb_EIO(smb_eio_trace_forced_shutdown); @@ -580,6 +581,7 @@ cifs_symlink(struct mnt_idmap *idmap, struct inode *inode, cifs_dbg(FYI, "symname is %s\n", symname); /* BB what if DFS and this volume is on different share? BB */ + sbflags = cifs_sb_flags(cifs_sb); rc = -EOPNOTSUPP; switch (cifs_symlink_type(cifs_sb)) { case CIFS_SYMLINK_TYPE_UNIX: @@ -594,14 +596,14 @@ cifs_symlink(struct mnt_idmap *idmap, struct inode *inode, break; case CIFS_SYMLINK_TYPE_MFSYMLINKS: - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) { + if (sbflags & CIFS_MOUNT_MF_SYMLINKS) { rc = create_mf_symlink(xid, pTcon, cifs_sb, full_path, symname); } break; case CIFS_SYMLINK_TYPE_SFU: - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + if (sbflags & CIFS_MOUNT_UNX_EMUL) { rc = __cifs_sfu_make_node(xid, inode, direntry, pTcon, full_path, S_IFLNK, 0, symname); diff --git a/fs/smb/client/misc.c b/fs/smb/client/misc.c index 22cde46309fe..2aff1cab6c31 100644 --- a/fs/smb/client/misc.c +++ b/fs/smb/client/misc.c @@ -28,6 +28,11 @@ #include "fs_context.h" #include "cached_dir.h" +struct tcon_list { + struct list_head entry; + struct cifs_tcon *tcon; +}; + /* The xid serves as a useful identifier for each incoming vfs request, in a similar way to the mid which is useful to track each sent smb, and CurrentXid can also provide a running counter (although it @@ -275,13 +280,15 @@ dump_smb(void *buf, int smb_buf_length) void cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) { + unsigned int sbflags = cifs_sb_flags(cifs_sb); + + if (sbflags & CIFS_MOUNT_SERVER_INUM) { struct cifs_tcon *tcon = NULL; if (cifs_sb->master_tlink) tcon = cifs_sb_master_tcon(cifs_sb); - cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM; + atomic_andnot(CIFS_MOUNT_SERVER_INUM, &cifs_sb->mnt_cifs_flags); cifs_sb->mnt_cifs_serverino_autodisabled = true; cifs_dbg(VFS, "Autodisabling the use of server inode numbers on %s\n", tcon ? tcon->tree_name : "new server"); @@ -382,11 +389,13 @@ void cifs_done_oplock_break(struct cifsInodeInfo *cinode) bool backup_cred(struct cifs_sb_info *cifs_sb) { - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_BACKUPUID) { + unsigned int sbflags = cifs_sb_flags(cifs_sb); + + if (sbflags & CIFS_MOUNT_CIFS_BACKUPUID) { if (uid_eq(cifs_sb->ctx->backupuid, current_fsuid())) return true; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_BACKUPGID) { + if (sbflags & CIFS_MOUNT_CIFS_BACKUPGID) { if (in_group_p(cifs_sb->ctx->backupgid)) return true; } @@ -550,6 +559,43 @@ cifs_close_all_deferred_files(struct cifs_tcon *tcon) } } +void cifs_close_all_deferred_files_sb(struct cifs_sb_info *cifs_sb) +{ + struct rb_root *root = &cifs_sb->tlink_tree; + struct rb_node *node; + struct cifs_tcon *tcon; + struct tcon_link *tlink; + struct tcon_list *tmp_list, *q; + LIST_HEAD(tcon_head); + + spin_lock(&cifs_sb->tlink_tree_lock); + for (node = rb_first(root); node; node = rb_next(node)) { + tlink = rb_entry(node, struct tcon_link, tl_rbnode); + tcon = tlink_tcon(tlink); + if (IS_ERR(tcon)) + continue; + tmp_list = kmalloc_obj(struct tcon_list, GFP_ATOMIC); + if (tmp_list == NULL) + break; + tmp_list->tcon = tcon; + /* Take a reference on tcon to prevent it from being freed */ + spin_lock(&tcon->tc_lock); + ++tcon->tc_count; + trace_smb3_tcon_ref(tcon->debug_id, tcon->tc_count, + netfs_trace_tcon_ref_get_close_defer_files); + spin_unlock(&tcon->tc_lock); + list_add_tail(&tmp_list->entry, &tcon_head); + } + spin_unlock(&cifs_sb->tlink_tree_lock); + + list_for_each_entry_safe(tmp_list, q, &tcon_head, entry) { + cifs_close_all_deferred_files(tmp_list->tcon); + list_del(&tmp_list->entry); + cifs_put_tcon(tmp_list->tcon, netfs_trace_tcon_ref_put_close_defer_files); + kfree(tmp_list); + } +} + void cifs_close_deferred_file_under_dentry(struct cifs_tcon *tcon, struct dentry *dentry) { @@ -955,7 +1001,7 @@ int cifs_update_super_prepath(struct cifs_sb_info *cifs_sb, char *prefix) convert_delimiter(cifs_sb->prepath, CIFS_DIR_SEP(cifs_sb)); } - cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + atomic_or(CIFS_MOUNT_USE_PREFIX_PATH, &cifs_sb->mnt_cifs_flags); return 0; } @@ -984,7 +1030,7 @@ int cifs_inval_name_dfs_link_error(const unsigned int xid, * look up or tcon is not DFS. */ if (strlen(full_path) < 2 || !cifs_sb || - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) || + (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_DFS) || !is_tcon_dfs(tcon)) return 0; diff --git a/fs/smb/client/readdir.c b/fs/smb/client/readdir.c index 8615a8747b7f..be22bbc4a65a 100644 --- a/fs/smb/client/readdir.c +++ b/fs/smb/client/readdir.c @@ -121,7 +121,7 @@ retry: * want to clobber the existing one with the one that * the readdir code created. */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_SERVER_INUM)) fattr->cf_uniqueid = CIFS_I(inode)->uniqueid; /* @@ -177,6 +177,7 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb) struct cifs_open_info_data data = { .reparse = { .tag = fattr->cf_cifstag, }, }; + unsigned int sbflags; fattr->cf_uid = cifs_sb->ctx->linux_uid; fattr->cf_gid = cifs_sb->ctx->linux_gid; @@ -215,12 +216,12 @@ out_reparse: * may look wrong since the inodes may not have timed out by the time * "ls" does a stat() call on them. */ - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) || - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID)) + sbflags = cifs_sb_flags(cifs_sb); + if (sbflags & (CIFS_MOUNT_CIFS_ACL | CIFS_MOUNT_MODE_FROM_SID)) fattr->cf_flags |= CIFS_FATTR_NEED_REVAL; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL && - fattr->cf_cifsattrs & ATTR_SYSTEM) { + if ((sbflags & CIFS_MOUNT_UNX_EMUL) && + (fattr->cf_cifsattrs & ATTR_SYSTEM)) { if (fattr->cf_eof == 0) { fattr->cf_mode &= ~S_IFMT; fattr->cf_mode |= S_IFIFO; @@ -345,13 +346,14 @@ static int _initiate_cifs_search(const unsigned int xid, struct file *file, const char *full_path) { - __u16 search_flags; - int rc = 0; - struct cifsFileInfo *cifsFile; - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file); + struct cifs_sb_info *cifs_sb = CIFS_SB(file); struct tcon_link *tlink = NULL; - struct cifs_tcon *tcon; struct TCP_Server_Info *server; + struct cifsFileInfo *cifsFile; + struct cifs_tcon *tcon; + unsigned int sbflags; + __u16 search_flags; + int rc = 0; if (file->private_data == NULL) { tlink = cifs_sb_tlink(cifs_sb); @@ -385,6 +387,7 @@ _initiate_cifs_search(const unsigned int xid, struct file *file, cifs_dbg(FYI, "Full path: %s start at: %lld\n", full_path, file->f_pos); ffirst_retry: + sbflags = cifs_sb_flags(cifs_sb); /* test for Unix extensions */ /* but now check for them on the share/mount not on the SMB session */ /* if (cap_unix(tcon->ses) { */ @@ -395,7 +398,7 @@ ffirst_retry: else if ((tcon->ses->capabilities & tcon->ses->server->vals->cap_nt_find) == 0) { cifsFile->srch_inf.info_level = SMB_FIND_FILE_INFO_STANDARD; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) { + } else if (sbflags & CIFS_MOUNT_SERVER_INUM) { cifsFile->srch_inf.info_level = SMB_FIND_FILE_ID_FULL_DIR_INFO; } else /* not srvinos - BB fixme add check for backlevel? */ { cifsFile->srch_inf.info_level = SMB_FIND_FILE_FULL_DIRECTORY_INFO; @@ -411,8 +414,7 @@ ffirst_retry: if (rc == 0) { cifsFile->invalidHandle = false; - } else if ((rc == -EOPNOTSUPP) && - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) { + } else if (rc == -EOPNOTSUPP && (sbflags & CIFS_MOUNT_SERVER_INUM)) { cifs_autodisable_serverino(cifs_sb); goto ffirst_retry; } @@ -690,7 +692,7 @@ find_cifs_entry(const unsigned int xid, struct cifs_tcon *tcon, loff_t pos, loff_t first_entry_in_buffer; loff_t index_to_find = pos; struct cifsFileInfo *cfile = file->private_data; - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file); + struct cifs_sb_info *cifs_sb = CIFS_SB(file); struct TCP_Server_Info *server = tcon->ses->server; /* check if index in the buffer */ @@ -955,6 +957,7 @@ static int cifs_filldir(char *find_entry, struct file *file, struct cifs_sb_info *cifs_sb = CIFS_SB(sb); struct cifs_dirent de = { NULL, }; struct cifs_fattr fattr; + unsigned int sbflags; struct qstr name; int rc = 0; @@ -1019,15 +1022,15 @@ static int cifs_filldir(char *find_entry, struct file *file, break; } - if (de.ino && (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) { + sbflags = cifs_sb_flags(cifs_sb); + if (de.ino && (sbflags & CIFS_MOUNT_SERVER_INUM)) { fattr.cf_uniqueid = de.ino; } else { fattr.cf_uniqueid = iunique(sb, ROOT_I); cifs_autodisable_serverino(cifs_sb); } - if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) && - couldbe_mf_symlink(&fattr)) + if ((sbflags & CIFS_MOUNT_MF_SYMLINKS) && couldbe_mf_symlink(&fattr)) /* * trying to get the type and mode can be slow, * so just call those regular files for now, and mark @@ -1058,7 +1061,7 @@ int cifs_readdir(struct file *file, struct dir_context *ctx) const char *full_path; void *page = alloc_dentry_path(); struct cached_fid *cfid = NULL; - struct cifs_sb_info *cifs_sb = CIFS_FILE_SB(file); + struct cifs_sb_info *cifs_sb = CIFS_SB(file); xid = get_xid(); diff --git a/fs/smb/client/reparse.c b/fs/smb/client/reparse.c index ce9b923498b5..cd1e1eaee67a 100644 --- a/fs/smb/client/reparse.c +++ b/fs/smb/client/reparse.c @@ -55,17 +55,18 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, const char *full_path, const char *symname) { struct reparse_symlink_data_buffer *buf = NULL; - struct cifs_open_info_data data = {}; - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); const char *symroot = cifs_sb->ctx->symlinkroot; + struct cifs_open_info_data data = {}; + char sep = CIFS_DIR_SEP(cifs_sb); + char *symlink_target = NULL; + u16 len, plen, poff, slen; + unsigned int sbflags; + __le16 *path = NULL; struct inode *new; + char *sym = NULL; struct kvec iov; - __le16 *path = NULL; bool directory; - char *symlink_target = NULL; - char *sym = NULL; - char sep = CIFS_DIR_SEP(cifs_sb); - u16 len, plen, poff, slen; int rc = 0; if (strlen(symname) > REPARSE_SYM_PATH_MAX) @@ -83,8 +84,8 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, .symlink_target = symlink_target, }; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && - symroot && symname[0] == '/') { + sbflags = cifs_sb_flags(cifs_sb); + if (!(sbflags & CIFS_MOUNT_POSIX_PATHS) && symroot && symname[0] == '/') { /* * This is a request to create an absolute symlink on the server * which does not support POSIX paths, and expects symlink in @@ -164,7 +165,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, * mask these characters in NT object prefix by '_' and then change * them back. */ - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') + if (!(sbflags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') sym[0] = sym[1] = sym[2] = sym[5] = '_'; path = cifs_convert_path_to_utf16(sym, cifs_sb); @@ -173,7 +174,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, goto out; } - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') { + if (!(sbflags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') { sym[0] = '\\'; sym[1] = sym[2] = '?'; sym[5] = ':'; @@ -197,7 +198,7 @@ static int create_native_symlink(const unsigned int xid, struct inode *inode, slen = 2 * UniStrnlen((wchar_t *)path, REPARSE_SYM_PATH_MAX); poff = 0; plen = slen; - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') { + if (!(sbflags & CIFS_MOUNT_POSIX_PATHS) && symname[0] == '/') { /* * For absolute NT symlinks skip leading "\\??\\" in PrintName as * PrintName is user visible location in DOS/Win32 format (not in NT format). @@ -824,7 +825,7 @@ int smb2_parse_native_symlink(char **target, const char *buf, unsigned int len, goto out; } - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) && + if (!(cifs_sb_flags(cifs_sb) & CIFS_MOUNT_POSIX_PATHS) && symroot && !relative) { /* * This is an absolute symlink from the server which does not diff --git a/fs/smb/client/reparse.h b/fs/smb/client/reparse.h index 570b0d25aeba..0164dc47bdfd 100644 --- a/fs/smb/client/reparse.h +++ b/fs/smb/client/reparse.h @@ -33,7 +33,7 @@ static inline kuid_t wsl_make_kuid(struct cifs_sb_info *cifs_sb, { u32 uid = le32_to_cpu(*(__le32 *)ptr); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_OVERR_UID) return cifs_sb->ctx->linux_uid; return make_kuid(current_user_ns(), uid); } @@ -43,7 +43,7 @@ static inline kgid_t wsl_make_kgid(struct cifs_sb_info *cifs_sb, { u32 gid = le32_to_cpu(*(__le32 *)ptr); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_OVERR_GID) return cifs_sb->ctx->linux_gid; return make_kgid(current_user_ns(), gid); } diff --git a/fs/smb/client/smb1encrypt.c b/fs/smb/client/smb1encrypt.c index 0dbbce2431ff..bf10fdeeedca 100644 --- a/fs/smb/client/smb1encrypt.c +++ b/fs/smb/client/smb1encrypt.c @@ -11,6 +11,7 @@ #include <linux/fips.h> #include <crypto/md5.h> +#include <crypto/utils.h> #include "cifsproto.h" #include "smb1proto.h" #include "cifs_debug.h" @@ -131,7 +132,7 @@ int cifs_verify_signature(struct smb_rqst *rqst, /* cifs_dump_mem("what we think it should be: ", what_we_think_sig_should_be, 16); */ - if (memcmp(server_response_sig, what_we_think_sig_should_be, 8)) + if (crypto_memneq(server_response_sig, what_we_think_sig_should_be, 8)) return -EACCES; else return 0; diff --git a/fs/smb/client/smb1ops.c b/fs/smb/client/smb1ops.c index aed49aaef8c4..9694117050a6 100644 --- a/fs/smb/client/smb1ops.c +++ b/fs/smb/client/smb1ops.c @@ -49,6 +49,7 @@ void reset_cifs_unix_caps(unsigned int xid, struct cifs_tcon *tcon, if (!CIFSSMBQFSUnixInfo(xid, tcon)) { __u64 cap = le64_to_cpu(tcon->fsUnixInfo.Capability); + unsigned int sbflags; cifs_dbg(FYI, "unix caps which server supports %lld\n", cap); /* @@ -75,14 +76,16 @@ void reset_cifs_unix_caps(unsigned int xid, struct cifs_tcon *tcon, if (cap & CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP) cifs_dbg(VFS, "per-share encryption not supported yet\n"); + if (cifs_sb) + sbflags = cifs_sb_flags(cifs_sb); + cap &= CIFS_UNIX_CAP_MASK; if (ctx && ctx->no_psx_acl) cap &= ~CIFS_UNIX_POSIX_ACL_CAP; else if (CIFS_UNIX_POSIX_ACL_CAP & cap) { cifs_dbg(FYI, "negotiated posix acl support\n"); if (cifs_sb) - cifs_sb->mnt_cifs_flags |= - CIFS_MOUNT_POSIXACL; + sbflags |= CIFS_MOUNT_POSIXACL; } if (ctx && ctx->posix_paths == 0) @@ -90,10 +93,12 @@ void reset_cifs_unix_caps(unsigned int xid, struct cifs_tcon *tcon, else if (cap & CIFS_UNIX_POSIX_PATHNAMES_CAP) { cifs_dbg(FYI, "negotiate posix pathnames\n"); if (cifs_sb) - cifs_sb->mnt_cifs_flags |= - CIFS_MOUNT_POSIX_PATHS; + sbflags |= CIFS_MOUNT_POSIX_PATHS; } + if (cifs_sb) + atomic_set(&cifs_sb->mnt_cifs_flags, sbflags); + cifs_dbg(FYI, "Negotiate caps 0x%x\n", (int)cap); #ifdef CONFIG_CIFS_DEBUG2 if (cap & CIFS_UNIX_FCNTL_CAP) @@ -955,7 +960,7 @@ smb_set_file_info(struct inode *inode, const char *full_path, struct cifs_tcon *tcon; /* if the file is already open for write, just use that fileid */ - open_file = find_writable_file(cinode, FIND_WR_FSUID_ONLY); + open_file = find_writable_file(cinode, FIND_FSUID_ONLY); if (open_file) { fid.netfid = open_file->fid.netfid; @@ -1147,7 +1152,7 @@ static int cifs_oplock_response(struct cifs_tcon *tcon, __u64 persistent_fid, __u64 volatile_fid, __u16 net_fid, struct cifsInodeInfo *cinode, unsigned int oplock) { - unsigned int sbflags = CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags; + unsigned int sbflags = cifs_sb_flags(CIFS_SB(cinode)); __u8 op; op = !!((oplock & CIFS_CACHE_READ_FLG) || (sbflags & CIFS_MOUNT_RO_CACHE)); @@ -1282,7 +1287,8 @@ cifs_make_node(unsigned int xid, struct inode *inode, struct dentry *dentry, struct cifs_tcon *tcon, const char *full_path, umode_t mode, dev_t dev) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode); + unsigned int sbflags = cifs_sb_flags(cifs_sb); struct inode *newinode = NULL; int rc; @@ -1298,7 +1304,7 @@ cifs_make_node(unsigned int xid, struct inode *inode, .mtime = NO_CHANGE_64, .device = dev, }; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { + if (sbflags & CIFS_MOUNT_SET_UID) { args.uid = current_fsuid(); args.gid = current_fsgid(); } else { @@ -1317,7 +1323,7 @@ cifs_make_node(unsigned int xid, struct inode *inode, if (rc == 0) d_instantiate(dentry, newinode); return rc; - } else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + } else if (sbflags & CIFS_MOUNT_UNX_EMUL) { /* * Check if mounted with mount parm 'sfu' mount parm. * SFU emulation should work with all servers diff --git a/fs/smb/client/smb1transport.c b/fs/smb/client/smb1transport.c index 38d6d5538b96..53abb29fe71b 100644 --- a/fs/smb/client/smb1transport.c +++ b/fs/smb/client/smb1transport.c @@ -460,7 +460,7 @@ check_smb_hdr(struct smb_hdr *smb) return 0; /* - * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING + * Windows NT server returns error response (e.g. STATUS_DELETE_PENDING * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) * for some TRANS2 requests without the RESPONSE flag set in header. */ diff --git a/fs/smb/client/smb2file.c b/fs/smb/client/smb2file.c index 1ab41de2b634..ed651c946251 100644 --- a/fs/smb/client/smb2file.c +++ b/fs/smb/client/smb2file.c @@ -72,7 +72,7 @@ int smb2_fix_symlink_target_type(char **target, bool directory, struct cifs_sb_i * POSIX server does not distinguish between symlinks to file and * symlink directory. So nothing is needed to fix on the client side. */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_POSIX_PATHS) return 0; if (!*target) diff --git a/fs/smb/client/smb2glob.h b/fs/smb/client/smb2glob.h index e56e4d402f13..19da74b1edab 100644 --- a/fs/smb/client/smb2glob.h +++ b/fs/smb/client/smb2glob.h @@ -46,4 +46,16 @@ enum smb2_compound_ops { #define END_OF_CHAIN 4 #define RELATED_REQUEST 8 +/* + ***************************************************************** + * Struct definitions go here + ***************************************************************** + */ + +struct status_to_posix_error { + __u32 smb2_status; + int posix_error; + char *status_string; +}; + #endif /* _SMB2_GLOB_H */ diff --git a/fs/smb/client/smb2inode.c b/fs/smb/client/smb2inode.c index 195a38fd61e8..364bdcff9c9d 100644 --- a/fs/smb/client/smb2inode.c +++ b/fs/smb/client/smb2inode.c @@ -325,7 +325,7 @@ replay_again: cfile->fid.volatile_fid, SMB_FIND_FILE_POSIX_INFO, SMB2_O_INFO_FILE, 0, - sizeof(struct smb311_posix_qinfo *) + + sizeof(struct smb311_posix_qinfo) + (PATH_MAX * 2) + (sizeof(struct smb_sid) * 2), 0, NULL); } else { @@ -335,7 +335,7 @@ replay_again: COMPOUND_FID, SMB_FIND_FILE_POSIX_INFO, SMB2_O_INFO_FILE, 0, - sizeof(struct smb311_posix_qinfo *) + + sizeof(struct smb311_posix_qinfo) + (PATH_MAX * 2) + (sizeof(struct smb_sid) * 2), 0, NULL); } @@ -1156,7 +1156,7 @@ smb2_mkdir_setinfo(struct inode *inode, const char *name, cifs_i = CIFS_I(inode); dosattrs = cifs_i->cifsAttrs | ATTR_READONLY; data.Attributes = cpu_to_le32(dosattrs); - cifs_get_writable_path(tcon, name, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, name, FIND_ANY, &cfile); oparms = CIFS_OPARMS(cifs_sb, tcon, name, FILE_WRITE_ATTRIBUTES, FILE_CREATE, CREATE_NOT_FILE, ACL_NO_MODE); tmprc = smb2_compound_op(xid, tcon, cifs_sb, name, @@ -1216,6 +1216,7 @@ again: memset(resp_buftype, 0, sizeof(resp_buftype)); memset(rsp_iov, 0, sizeof(rsp_iov)); + memset(open_iov, 0, sizeof(open_iov)); rqst[0].rq_iov = open_iov; rqst[0].rq_nvec = ARRAY_SIZE(open_iov); @@ -1240,14 +1241,15 @@ again: creq = rqst[0].rq_iov[0].iov_base; creq->ShareAccess = FILE_SHARE_DELETE_LE; + memset(&close_iov, 0, sizeof(close_iov)); rqst[1].rq_iov = &close_iov; rqst[1].rq_nvec = 1; rc = SMB2_close_init(tcon, server, &rqst[1], COMPOUND_FID, COMPOUND_FID, false); - smb2_set_related(&rqst[1]); if (rc) goto err_free; + smb2_set_related(&rqst[1]); if (retries) { /* Back-off before retry */ @@ -1334,14 +1336,13 @@ int smb2_rename_path(const unsigned int xid, __u32 co = file_create_options(source_dentry); drop_cached_dir_by_name(xid, tcon, from_name, cifs_sb); - cifs_get_writable_path(tcon, from_name, FIND_WR_WITH_DELETE, &cfile); + cifs_get_writable_path(tcon, from_name, FIND_WITH_DELETE, &cfile); int rc = smb2_set_path_attr(xid, tcon, from_name, to_name, cifs_sb, co, DELETE, SMB2_OP_RENAME, cfile, source_dentry); if (rc == -EINVAL) { cifs_dbg(FYI, "invalid lease key, resending request without lease"); - cifs_get_writable_path(tcon, from_name, - FIND_WR_WITH_DELETE, &cfile); + cifs_get_writable_path(tcon, from_name, FIND_WITH_DELETE, &cfile); rc = smb2_set_path_attr(xid, tcon, from_name, to_name, cifs_sb, co, DELETE, SMB2_OP_RENAME, cfile, NULL); } @@ -1375,7 +1376,7 @@ smb2_set_path_size(const unsigned int xid, struct cifs_tcon *tcon, in_iov.iov_base = &eof; in_iov.iov_len = sizeof(eof); - cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_ANY, &cfile); oparms = CIFS_OPARMS(cifs_sb, tcon, full_path, FILE_WRITE_DATA, FILE_OPEN, 0, ACL_NO_MODE); @@ -1385,7 +1386,7 @@ smb2_set_path_size(const unsigned int xid, struct cifs_tcon *tcon, cfile, NULL, NULL, dentry); if (rc == -EINVAL) { cifs_dbg(FYI, "invalid lease key, resending request without lease"); - cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_ANY, &cfile); rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, &in_iov, &(int){SMB2_OP_SET_EOF}, 1, @@ -1415,7 +1416,7 @@ smb2_set_file_info(struct inode *inode, const char *full_path, (buf->LastWriteTime == 0) && (buf->ChangeTime == 0)) { if (buf->Attributes == 0) goto out; /* would be a no op, no sense sending this */ - cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_ANY, &cfile); } oparms = CIFS_OPARMS(cifs_sb, tcon, full_path, FILE_WRITE_ATTRIBUTES, @@ -1474,7 +1475,7 @@ struct inode *smb2_create_reparse_inode(struct cifs_open_info_data *data, if (tcon->posix_extensions) { cmds[1] = SMB2_OP_POSIX_QUERY_INFO; - cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_ANY, &cfile); rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, in_iov, cmds, 2, cfile, out_iov, out_buftype, NULL); if (!rc) { @@ -1483,7 +1484,7 @@ struct inode *smb2_create_reparse_inode(struct cifs_open_info_data *data, } } else { cmds[1] = SMB2_OP_QUERY_INFO; - cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_ANY, &cfile); rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, in_iov, cmds, 2, cfile, out_iov, out_buftype, NULL); if (!rc) { @@ -1634,13 +1635,12 @@ int smb2_rename_pending_delete(const char *full_path, iov[1].iov_base = utf16_path; iov[1].iov_len = sizeof(*utf16_path) * UniStrlen((wchar_t *)utf16_path); - cifs_get_writable_path(tcon, full_path, FIND_WR_WITH_DELETE, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_WITH_DELETE, &cfile); rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, iov, cmds, num_cmds, cfile, NULL, NULL, dentry); if (rc == -EINVAL) { cifs_dbg(FYI, "invalid lease key, resending request without lease\n"); - cifs_get_writable_path(tcon, full_path, - FIND_WR_WITH_DELETE, &cfile); + cifs_get_writable_path(tcon, full_path, FIND_WITH_DELETE, &cfile); rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, &oparms, iov, cmds, num_cmds, cfile, NULL, NULL, NULL); } diff --git a/fs/smb/client/smb2maperror.c b/fs/smb/client/smb2maperror.c index cd036365201f..2b8782c4f684 100644 --- a/fs/smb/client/smb2maperror.c +++ b/fs/smb/client/smb2maperror.c @@ -8,7 +8,6 @@ * */ #include <linux/errno.h> -#include "cifsglob.h" #include "cifsproto.h" #include "cifs_debug.h" #include "smb2proto.h" @@ -16,12 +15,6 @@ #include "../common/smb2status.h" #include "trace.h" -struct status_to_posix_error { - __u32 smb2_status; - int posix_error; - char *status_string; -}; - static const struct status_to_posix_error smb2_error_map_table[] = { /* * Automatically generated by the `gen_smb2_mapping` script, @@ -115,10 +108,22 @@ int __init smb2_init_maperror(void) return 0; } -#define SMB_CLIENT_KUNIT_AVAILABLE \ - ((IS_MODULE(CONFIG_CIFS) && IS_ENABLED(CONFIG_KUNIT)) || \ - (IS_BUILTIN(CONFIG_CIFS) && IS_BUILTIN(CONFIG_KUNIT))) +#if IS_ENABLED(CONFIG_SMB_KUNIT_TESTS) +#define EXPORT_SYMBOL_FOR_SMB_TEST(sym) \ + EXPORT_SYMBOL_FOR_MODULES(sym, "smb2maperror_test") + +/* Previous prototype for eliminating the build warning. */ +const struct status_to_posix_error *smb2_get_err_map_test(__u32 smb2_status); + +const struct status_to_posix_error *smb2_get_err_map_test(__u32 smb2_status) +{ + return smb2_get_err_map(smb2_status); +} +EXPORT_SYMBOL_FOR_SMB_TEST(smb2_get_err_map_test); + +const struct status_to_posix_error *smb2_error_map_table_test = smb2_error_map_table; +EXPORT_SYMBOL_FOR_SMB_TEST(smb2_error_map_table_test); -#if SMB_CLIENT_KUNIT_AVAILABLE && IS_ENABLED(CONFIG_SMB_KUNIT_TESTS) -#include "smb2maperror_test.c" -#endif /* CONFIG_SMB_KUNIT_TESTS */ +unsigned int smb2_error_map_num = ARRAY_SIZE(smb2_error_map_table); +EXPORT_SYMBOL_FOR_SMB_TEST(smb2_error_map_num); +#endif diff --git a/fs/smb/client/smb2maperror_test.c b/fs/smb/client/smb2maperror_test.c index 38ea6b846a99..8c47dea7a2c1 100644 --- a/fs/smb/client/smb2maperror_test.c +++ b/fs/smb/client/smb2maperror_test.c @@ -9,13 +9,18 @@ */ #include <kunit/test.h> +#include "smb2glob.h" + +const struct status_to_posix_error *smb2_get_err_map_test(__u32 smb2_status); +extern const struct status_to_posix_error *smb2_error_map_table_test; +extern unsigned int smb2_error_map_num; static void test_cmp_map(struct kunit *test, const struct status_to_posix_error *expect) { const struct status_to_posix_error *result; - result = smb2_get_err_map(expect->smb2_status); + result = smb2_get_err_map_test(expect->smb2_status); KUNIT_EXPECT_PTR_NE(test, NULL, result); KUNIT_EXPECT_EQ(test, expect->smb2_status, result->smb2_status); KUNIT_EXPECT_EQ(test, expect->posix_error, result->posix_error); @@ -26,8 +31,8 @@ static void maperror_test_check_search(struct kunit *test) { unsigned int i; - for (i = 0; i < ARRAY_SIZE(smb2_error_map_table); i++) - test_cmp_map(test, &smb2_error_map_table[i]); + for (i = 0; i < smb2_error_map_num; i++) + test_cmp_map(test, &smb2_error_map_table_test[i]); } static struct kunit_case maperror_test_cases[] = { @@ -43,3 +48,4 @@ static struct kunit_suite maperror_suite = { kunit_test_suite(maperror_suite); MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("KUnit tests of SMB2 maperror"); diff --git a/fs/smb/client/smb2misc.c b/fs/smb/client/smb2misc.c index e19674d9b92b..973fce3c959c 100644 --- a/fs/smb/client/smb2misc.c +++ b/fs/smb/client/smb2misc.c @@ -455,17 +455,8 @@ calc_size_exit: __le16 * cifs_convert_path_to_utf16(const char *from, struct cifs_sb_info *cifs_sb) { - int len; const char *start_of_path; - __le16 *to; - int map_type; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SFM_CHR) - map_type = SFM_MAP_UNI_RSVD; - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR) - map_type = SFU_MAP_UNI_RSVD; - else - map_type = NO_MAP_UNI_RSVD; + int len; /* Windows doesn't allow paths beginning with \ */ if (from[0] == '\\') @@ -479,14 +470,13 @@ cifs_convert_path_to_utf16(const char *from, struct cifs_sb_info *cifs_sb) } else start_of_path = from; - to = cifs_strndup_to_utf16(start_of_path, PATH_MAX, &len, - cifs_sb->local_nls, map_type); - return to; + return cifs_strndup_to_utf16(start_of_path, PATH_MAX, &len, + cifs_sb->local_nls, cifs_remap(cifs_sb)); } __le32 smb2_get_lease_state(struct cifsInodeInfo *cinode, unsigned int oplock) { - unsigned int sbflags = CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags; + unsigned int sbflags = cifs_sb_flags(CIFS_SB(cinode)); __le32 lease = 0; if ((oplock & CIFS_CACHE_WRITE_FLG) || (sbflags & CIFS_MOUNT_RW_CACHE)) diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index fea9a35caa57..509fcea28a42 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -628,6 +628,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, struct smb_sockaddr_in6 *p6; struct cifs_server_iface *info = NULL, *iface = NULL, *niface = NULL; struct cifs_server_iface tmp_iface; + __be16 port; ssize_t bytes_left; size_t next = 0; int nb_iface = 0; @@ -662,6 +663,15 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, goto out; } + spin_lock(&ses->server->srv_lock); + if (ses->server->dstaddr.ss_family == AF_INET) + port = ((struct sockaddr_in *)&ses->server->dstaddr)->sin_port; + else if (ses->server->dstaddr.ss_family == AF_INET6) + port = ((struct sockaddr_in6 *)&ses->server->dstaddr)->sin6_port; + else + port = cpu_to_be16(CIFS_PORT); + spin_unlock(&ses->server->srv_lock); + while (bytes_left >= (ssize_t)sizeof(*p)) { memset(&tmp_iface, 0, sizeof(tmp_iface)); /* default to 1Gbps when link speed is unset */ @@ -682,7 +692,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, memcpy(&addr4->sin_addr, &p4->IPv4Address, 4); /* [MS-SMB2] 2.2.32.5.1.1 Clients MUST ignore these */ - addr4->sin_port = cpu_to_be16(CIFS_PORT); + addr4->sin_port = port; cifs_dbg(FYI, "%s: ipv4 %pI4\n", __func__, &addr4->sin_addr); @@ -696,7 +706,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, /* [MS-SMB2] 2.2.32.5.1.2 Clients MUST ignore these */ addr6->sin6_flowinfo = 0; addr6->sin6_scope_id = 0; - addr6->sin6_port = cpu_to_be16(CIFS_PORT); + addr6->sin6_port = port; cifs_dbg(FYI, "%s: ipv6 %pI6\n", __func__, &addr6->sin6_addr); @@ -986,7 +996,7 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, rc = -EREMOTE; } if (rc == -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && - (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS)) + (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_DFS)) rc = -EOPNOTSUPP; goto out; } @@ -1487,6 +1497,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, { struct smb2_file_network_open_info file_inf; struct inode *inode; + u64 asize; int rc; rc = __SMB2_close(xid, tcon, cfile->fid.persistent_fid, @@ -1510,14 +1521,9 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, inode_set_atime_to_ts(inode, cifs_NTtimeToUnix(file_inf.LastAccessTime)); - /* - * i_blocks is not related to (i_size / i_blksize), - * but instead 512 byte (2**9) size is required for - * calculating num blocks. - */ - if (le64_to_cpu(file_inf.AllocationSize) > 4096) - inode->i_blocks = - (512 - 1 + le64_to_cpu(file_inf.AllocationSize)) >> 9; + asize = le64_to_cpu(file_inf.AllocationSize); + if (asize > 4096) + inode->i_blocks = CIFS_INO_BLOCKS(asize); /* End of file and Attributes should not have to be updated on close */ spin_unlock(&inode->i_lock); @@ -2194,14 +2200,6 @@ smb2_duplicate_extents(const unsigned int xid, rc = smb2_set_file_size(xid, tcon, trgtfile, dest_off + len, false); if (rc) goto duplicate_extents_out; - - /* - * Although also could set plausible allocation size (i_blocks) - * here in addition to setting the file size, in reflink - * it is likely that the target file is sparse. Its allocation - * size will be queried on next revalidate, but it is important - * to make sure that file's cached size is updated immediately - */ netfs_resize_file(netfs_inode(inode), dest_off + len, true); cifs_setsize(inode, dest_off + len); } @@ -2691,7 +2689,7 @@ static int smb2_oplock_response(struct cifs_tcon *tcon, __u64 persistent_fid, __u64 volatile_fid, __u16 net_fid, struct cifsInodeInfo *cinode, unsigned int oplock) { - unsigned int sbflags = CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags; + unsigned int sbflags = cifs_sb_flags(CIFS_SB(cinode)); __u8 op; if (tcon->ses->server->capabilities & SMB2_GLOBAL_CAP_LEASING) @@ -3352,7 +3350,7 @@ get_smb2_acl(struct cifs_sb_info *cifs_sb, struct cifsFileInfo *open_file = NULL; if (inode && !(info & SACL_SECINFO)) - open_file = find_readable_file(CIFS_I(inode), true); + open_file = find_readable_file(CIFS_I(inode), FIND_FSUID_ONLY); if (!open_file || (info & SACL_SECINFO)) return get_smb2_acl_by_path(cifs_sb, path, pacllen, info); @@ -3898,7 +3896,7 @@ static loff_t smb3_llseek(struct file *file, struct cifs_tcon *tcon, loff_t offs * some servers (Windows2016) will not reflect recent writes in * QUERY_ALLOCATED_RANGES until SMB2_flush is called. */ - wrcfile = find_writable_file(cifsi, FIND_WR_ANY); + wrcfile = find_writable_file(cifsi, FIND_ANY); if (wrcfile) { filemap_write_and_wait(inode->i_mapping); smb2_flush_file(xid, tcon, &wrcfile->fid); @@ -5332,7 +5330,7 @@ static int smb2_make_node(unsigned int xid, struct inode *inode, struct dentry *dentry, struct cifs_tcon *tcon, const char *full_path, umode_t mode, dev_t dev) { - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + unsigned int sbflags = cifs_sb_flags(CIFS_SB(inode)); int rc = -EOPNOTSUPP; /* @@ -5341,7 +5339,7 @@ static int smb2_make_node(unsigned int xid, struct inode *inode, * supports block and char device, socket & fifo, * and was used by default in earlier versions of Windows */ - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + if (sbflags & CIFS_MOUNT_UNX_EMUL) { rc = cifs_sfu_make_node(xid, inode, dentry, tcon, full_path, mode, dev); } else if (CIFS_REPARSE_SUPPORT(tcon)) { diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index ef655acf673d..5188218c25be 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -1714,19 +1714,17 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data) is_binding = (ses->ses_status == SES_GOOD); spin_unlock(&ses->ses_lock); - /* keep session key if binding */ - if (!is_binding) { - kfree_sensitive(ses->auth_key.response); - ses->auth_key.response = kmemdup(msg->data, msg->sesskey_len, - GFP_KERNEL); - if (!ses->auth_key.response) { - cifs_dbg(VFS, "Kerberos can't allocate (%u bytes) memory\n", - msg->sesskey_len); - rc = -ENOMEM; - goto out_put_spnego_key; - } - ses->auth_key.len = msg->sesskey_len; + kfree_sensitive(ses->auth_key.response); + ses->auth_key.response = kmemdup(msg->data, + msg->sesskey_len, + GFP_KERNEL); + if (!ses->auth_key.response) { + cifs_dbg(VFS, "%s: can't allocate (%u bytes) memory\n", + __func__, msg->sesskey_len); + rc = -ENOMEM; + goto out_put_spnego_key; } + ses->auth_key.len = msg->sesskey_len; sess_data->iov[1].iov_base = msg->data + msg->sesskey_len; sess_data->iov[1].iov_len = msg->secblob_len; @@ -3182,22 +3180,19 @@ SMB2_open_init(struct cifs_tcon *tcon, struct TCP_Server_Info *server, } if ((oparms->disposition != FILE_OPEN) && (oparms->cifs_sb)) { + unsigned int sbflags = cifs_sb_flags(oparms->cifs_sb); bool set_mode; bool set_owner; - if ((oparms->cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) && - (oparms->mode != ACL_NO_MODE)) + if ((sbflags & CIFS_MOUNT_MODE_FROM_SID) && + oparms->mode != ACL_NO_MODE) { set_mode = true; - else { + } else { set_mode = false; oparms->mode = ACL_NO_MODE; } - if (oparms->cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) - set_owner = true; - else - set_owner = false; - + set_owner = sbflags & CIFS_MOUNT_UID_FROM_ACL; if (set_owner | set_mode) { cifs_dbg(FYI, "add sd with mode 0x%x\n", oparms->mode); rc = add_sd_context(iov, &n_iov, oparms->mode, set_owner); @@ -3994,24 +3989,6 @@ int SMB2_query_info(const unsigned int xid, struct cifs_tcon *tcon, NULL); } -#if 0 -/* currently unused, as now we are doing compounding instead (see smb311_posix_query_path_info) */ -int -SMB311_posix_query_info(const unsigned int xid, struct cifs_tcon *tcon, - u64 persistent_fid, u64 volatile_fid, - struct smb311_posix_qinfo *data, u32 *plen) -{ - size_t output_len = sizeof(struct smb311_posix_qinfo *) + - (sizeof(struct smb_sid) * 2) + (PATH_MAX * 2); - *plen = 0; - - return query_info(xid, tcon, persistent_fid, volatile_fid, - SMB_FIND_FILE_POSIX_INFO, SMB2_O_INFO_FILE, 0, - output_len, sizeof(struct smb311_posix_qinfo), (void **)&data, plen); - /* Note caller must free "data" (passed in above). It may be allocated in query_info call */ -} -#endif - int SMB2_query_acl(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, u64 volatile_fid, @@ -5330,7 +5307,10 @@ replay_again: memset(&rqst, 0, sizeof(struct smb_rqst)); rqst.rq_iov = iov; - rqst.rq_nvec = n_vec + 1; + /* iov[0] is the SMB header; move payload to rq_iter for encryption safety */ + rqst.rq_nvec = 1; + iov_iter_kvec(&rqst.rq_iter, ITER_SOURCE, &iov[1], n_vec, + io_parms->length); if (retries) { /* Back-off before retry */ diff --git a/fs/smb/client/smb2pdu.h b/fs/smb/client/smb2pdu.h index 78bb99f29d38..30d70097fe2f 100644 --- a/fs/smb/client/smb2pdu.h +++ b/fs/smb/client/smb2pdu.h @@ -224,7 +224,7 @@ struct smb2_file_reparse_point_info { __le32 Tag; } __packed; -/* See MS-FSCC 2.4.21 */ +/* See MS-FSCC 2.4.26 */ struct smb2_file_id_information { __le64 VolumeSerialNumber; __u64 PersistentFileId; /* opaque endianness */ @@ -251,7 +251,10 @@ struct smb2_file_id_extd_directory_info { extern char smb2_padding[7]; -/* equivalent of the contents of SMB3.1.1 POSIX open context response */ +/* + * See POSIX-SMB2 2.2.14.2.16 + * Link: https://gitlab.com/samba-team/smb3-posix-spec/-/blob/master/smb3_posix_extensions.md + */ struct create_posix_rsp { u32 nlink; u32 reparse_tag; diff --git a/fs/smb/client/smb2proto.h b/fs/smb/client/smb2proto.h index 881e42cf66ce..230bb1e9f4e1 100644 --- a/fs/smb/client/smb2proto.h +++ b/fs/smb/client/smb2proto.h @@ -167,9 +167,6 @@ int SMB2_flush_init(const unsigned int xid, struct smb_rqst *rqst, struct cifs_tcon *tcon, struct TCP_Server_Info *server, u64 persistent_fid, u64 volatile_fid); void SMB2_flush_free(struct smb_rqst *rqst); -int SMB311_posix_query_info(const unsigned int xid, struct cifs_tcon *tcon, - u64 persistent_fid, u64 volatile_fid, - struct smb311_posix_qinfo *data, u32 *plen); int SMB2_query_info(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, u64 volatile_fid, struct smb2_file_all_info *data); diff --git a/fs/smb/client/smb2transport.c b/fs/smb/client/smb2transport.c index 8b9000a83181..81be2b226e26 100644 --- a/fs/smb/client/smb2transport.c +++ b/fs/smb/client/smb2transport.c @@ -20,6 +20,7 @@ #include <linux/highmem.h> #include <crypto/aead.h> #include <crypto/sha2.h> +#include <crypto/utils.h> #include "cifsglob.h" #include "cifsproto.h" #include "smb2proto.h" @@ -617,7 +618,8 @@ smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server) if (rc) return rc; - if (memcmp(server_response_sig, shdr->Signature, SMB2_SIGNATURE_SIZE)) { + if (crypto_memneq(server_response_sig, shdr->Signature, + SMB2_SIGNATURE_SIZE)) { cifs_dbg(VFS, "sign fail cmd 0x%x message id 0x%llx\n", shdr->Command, shdr->MessageId); return -EACCES; diff --git a/fs/smb/client/trace.h b/fs/smb/client/trace.h index 9228f95cae2b..acfbb63086ea 100644 --- a/fs/smb/client/trace.h +++ b/fs/smb/client/trace.h @@ -176,6 +176,7 @@ EM(netfs_trace_tcon_ref_get_cached_laundromat, "GET Ch-Lau") \ EM(netfs_trace_tcon_ref_get_cached_lease_break, "GET Ch-Lea") \ EM(netfs_trace_tcon_ref_get_cancelled_close, "GET Cn-Cls") \ + EM(netfs_trace_tcon_ref_get_close_defer_files, "GET Cl-Def") \ EM(netfs_trace_tcon_ref_get_dfs_refer, "GET DfsRef") \ EM(netfs_trace_tcon_ref_get_find, "GET Find ") \ EM(netfs_trace_tcon_ref_get_find_sess_tcon, "GET FndSes") \ @@ -187,6 +188,7 @@ EM(netfs_trace_tcon_ref_put_cancelled_close, "PUT Cn-Cls") \ EM(netfs_trace_tcon_ref_put_cancelled_close_fid, "PUT Cn-Fid") \ EM(netfs_trace_tcon_ref_put_cancelled_mid, "PUT Cn-Mid") \ + EM(netfs_trace_tcon_ref_put_close_defer_files, "PUT Cl-Def") \ EM(netfs_trace_tcon_ref_put_mnt_ctx, "PUT MntCtx") \ EM(netfs_trace_tcon_ref_put_dfs_refer, "PUT DfsRfr") \ EM(netfs_trace_tcon_ref_put_reconnect_server, "PUT Reconn") \ diff --git a/fs/smb/client/transport.c b/fs/smb/client/transport.c index 75697f6d2566..05f8099047e1 100644 --- a/fs/smb/client/transport.c +++ b/fs/smb/client/transport.c @@ -807,16 +807,21 @@ cifs_cancelled_callback(struct TCP_Server_Info *server, struct mid_q_entry *mid) } /* - * Return a channel (master if none) of @ses that can be used to send - * regular requests. + * cifs_pick_channel - pick an eligible channel for network operations * - * If we are currently binding a new channel (negprot/sess.setup), - * return the new incomplete channel. + * @ses: session reference + * + * Select an eligible channel (not terminating and not marked as needing + * reconnect), preferring the least loaded one. If no eligible channel is + * found, fall back to the primary channel (index 0). + * + * Return: TCP_Server_Info pointer for the chosen channel, or NULL if @ses is + * NULL. */ struct TCP_Server_Info *cifs_pick_channel(struct cifs_ses *ses) { uint index = 0; - unsigned int min_in_flight = UINT_MAX, max_in_flight = 0; + unsigned int min_in_flight = UINT_MAX; struct TCP_Server_Info *server = NULL; int i, start, cur; @@ -846,14 +851,8 @@ struct TCP_Server_Info *cifs_pick_channel(struct cifs_ses *ses) min_in_flight = server->in_flight; index = cur; } - if (server->in_flight > max_in_flight) - max_in_flight = server->in_flight; } - /* if all channels are equally loaded, fall back to round-robin */ - if (min_in_flight == max_in_flight) - index = (uint)start % ses->chan_count; - server = ses->chans[index].server; spin_unlock(&ses->chan_lock); diff --git a/fs/smb/client/xattr.c b/fs/smb/client/xattr.c index e1a7d9a10a53..23227f2f9428 100644 --- a/fs/smb/client/xattr.c +++ b/fs/smb/client/xattr.c @@ -149,7 +149,7 @@ static int cifs_xattr_set(const struct xattr_handler *handler, break; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_XATTR) goto out; if (pTcon->ses->server->ops->set_EA) { @@ -309,7 +309,7 @@ static int cifs_xattr_get(const struct xattr_handler *handler, break; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_XATTR) goto out; if (pTcon->ses->server->ops->query_all_EAs) @@ -398,7 +398,7 @@ ssize_t cifs_listxattr(struct dentry *direntry, char *data, size_t buf_size) if (unlikely(cifs_forced_shutdown(cifs_sb))) return smb_EIO(smb_eio_trace_forced_shutdown); - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR) + if (cifs_sb_flags(cifs_sb) & CIFS_MOUNT_NO_XATTR) return -EOPNOTSUPP; tlink = cifs_sb_tlink(cifs_sb); diff --git a/fs/smb/server/Kconfig b/fs/smb/server/Kconfig index 2775162c535c..12594879cb64 100644 --- a/fs/smb/server/Kconfig +++ b/fs/smb/server/Kconfig @@ -13,6 +13,7 @@ config SMB_SERVER select CRYPTO_LIB_MD5 select CRYPTO_LIB_SHA256 select CRYPTO_LIB_SHA512 + select CRYPTO_LIB_UTILS select CRYPTO_CMAC select CRYPTO_AEAD2 select CRYPTO_CCM diff --git a/fs/smb/server/auth.c b/fs/smb/server/auth.c index 580c4d303dc3..af5f40304331 100644 --- a/fs/smb/server/auth.c +++ b/fs/smb/server/auth.c @@ -15,6 +15,7 @@ #include <crypto/aead.h> #include <crypto/md5.h> #include <crypto/sha2.h> +#include <crypto/utils.h> #include <linux/random.h> #include <linux/scatterlist.h> @@ -165,7 +166,8 @@ int ksmbd_auth_ntlmv2(struct ksmbd_conn *conn, struct ksmbd_session *sess, ntlmv2_rsp, CIFS_HMAC_MD5_HASH_SIZE, sess->sess_key); - if (memcmp(ntlmv2->ntlmv2_hash, ntlmv2_rsp, CIFS_HMAC_MD5_HASH_SIZE) != 0) + if (crypto_memneq(ntlmv2->ntlmv2_hash, ntlmv2_rsp, + CIFS_HMAC_MD5_HASH_SIZE)) return -EINVAL; return 0; } @@ -587,12 +589,8 @@ static int generate_smb3signingkey(struct ksmbd_session *sess, if (!(conn->dialect >= SMB30_PROT_ID && signing->binding)) memcpy(chann->smb3signingkey, key, SMB3_SIGN_KEY_SIZE); - ksmbd_debug(AUTH, "dumping generated AES signing keys\n"); + ksmbd_debug(AUTH, "generated SMB3 signing key\n"); ksmbd_debug(AUTH, "Session Id %llu\n", sess->id); - ksmbd_debug(AUTH, "Session Key %*ph\n", - SMB2_NTLMV2_SESSKEY_SIZE, sess->sess_key); - ksmbd_debug(AUTH, "Signing Key %*ph\n", - SMB3_SIGN_KEY_SIZE, key); return 0; } @@ -650,23 +648,9 @@ static void generate_smb3encryptionkey(struct ksmbd_conn *conn, ptwin->decryption.context, sess->smb3decryptionkey, SMB3_ENC_DEC_KEY_SIZE); - ksmbd_debug(AUTH, "dumping generated AES encryption keys\n"); + ksmbd_debug(AUTH, "generated SMB3 encryption/decryption keys\n"); ksmbd_debug(AUTH, "Cipher type %d\n", conn->cipher_type); ksmbd_debug(AUTH, "Session Id %llu\n", sess->id); - ksmbd_debug(AUTH, "Session Key %*ph\n", - SMB2_NTLMV2_SESSKEY_SIZE, sess->sess_key); - if (conn->cipher_type == SMB2_ENCRYPTION_AES256_CCM || - conn->cipher_type == SMB2_ENCRYPTION_AES256_GCM) { - ksmbd_debug(AUTH, "ServerIn Key %*ph\n", - SMB3_GCM256_CRYPTKEY_SIZE, sess->smb3encryptionkey); - ksmbd_debug(AUTH, "ServerOut Key %*ph\n", - SMB3_GCM256_CRYPTKEY_SIZE, sess->smb3decryptionkey); - } else { - ksmbd_debug(AUTH, "ServerIn Key %*ph\n", - SMB3_GCM128_CRYPTKEY_SIZE, sess->smb3encryptionkey); - ksmbd_debug(AUTH, "ServerOut Key %*ph\n", - SMB3_GCM128_CRYPTKEY_SIZE, sess->smb3decryptionkey); - } } void ksmbd_gen_smb30_encryptionkey(struct ksmbd_conn *conn, diff --git a/fs/smb/server/mgmt/tree_connect.c b/fs/smb/server/mgmt/tree_connect.c index a72d7e42a6c2..58e5b8592da4 100644 --- a/fs/smb/server/mgmt/tree_connect.c +++ b/fs/smb/server/mgmt/tree_connect.c @@ -102,8 +102,10 @@ out_error: void ksmbd_tree_connect_put(struct ksmbd_tree_connect *tcon) { - if (atomic_dec_and_test(&tcon->refcount)) + if (atomic_dec_and_test(&tcon->refcount)) { + ksmbd_share_config_put(tcon->share_conf); kfree(tcon); + } } static int __ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, @@ -113,10 +115,11 @@ static int __ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, ret = ksmbd_ipc_tree_disconnect_request(sess->id, tree_conn->id); ksmbd_release_tree_conn_id(sess, tree_conn->id); - ksmbd_share_config_put(tree_conn->share_conf); ksmbd_counter_dec(KSMBD_COUNTER_TREE_CONNS); - if (atomic_dec_and_test(&tree_conn->refcount)) + if (atomic_dec_and_test(&tree_conn->refcount)) { + ksmbd_share_config_put(tree_conn->share_conf); kfree(tree_conn); + } return ret; } diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index 09d9878db9cb..393a4ae47cc1 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -120,7 +120,7 @@ static void free_lease(struct oplock_info *opinfo) kfree(lease); } -static void free_opinfo(struct oplock_info *opinfo) +static void __free_opinfo(struct oplock_info *opinfo) { if (opinfo->is_lease) free_lease(opinfo); @@ -129,6 +129,18 @@ static void free_opinfo(struct oplock_info *opinfo) kfree(opinfo); } +static void free_opinfo_rcu(struct rcu_head *rcu) +{ + struct oplock_info *opinfo = container_of(rcu, struct oplock_info, rcu); + + __free_opinfo(opinfo); +} + +static void free_opinfo(struct oplock_info *opinfo) +{ + call_rcu(&opinfo->rcu, free_opinfo_rcu); +} + struct oplock_info *opinfo_get(struct ksmbd_file *fp) { struct oplock_info *opinfo; @@ -176,9 +188,9 @@ void opinfo_put(struct oplock_info *opinfo) free_opinfo(opinfo); } -static void opinfo_add(struct oplock_info *opinfo) +static void opinfo_add(struct oplock_info *opinfo, struct ksmbd_file *fp) { - struct ksmbd_inode *ci = opinfo->o_fp->f_ci; + struct ksmbd_inode *ci = fp->f_ci; down_write(&ci->m_lock); list_add(&opinfo->op_entry, &ci->m_op_list); @@ -1123,10 +1135,12 @@ void smb_lazy_parent_lease_break_close(struct ksmbd_file *fp) rcu_read_lock(); opinfo = rcu_dereference(fp->f_opinfo); - rcu_read_unlock(); - if (!opinfo || !opinfo->is_lease || opinfo->o_lease->version != 2) + if (!opinfo || !opinfo->is_lease || opinfo->o_lease->version != 2) { + rcu_read_unlock(); return; + } + rcu_read_unlock(); p_ci = ksmbd_inode_lookup_lock(fp->filp->f_path.dentry->d_parent); if (!p_ci) @@ -1277,20 +1291,21 @@ set_lev: set_oplock_level(opinfo, req_op_level, lctx); out: - rcu_assign_pointer(fp->f_opinfo, opinfo); - opinfo->o_fp = fp; - opinfo_count_inc(fp); - opinfo_add(opinfo); + opinfo_add(opinfo, fp); + if (opinfo->is_lease) { err = add_lease_global_list(opinfo); if (err) goto err_out; } + rcu_assign_pointer(fp->f_opinfo, opinfo); + opinfo->o_fp = fp; + return 0; err_out: - free_opinfo(opinfo); + __free_opinfo(opinfo); return err; } diff --git a/fs/smb/server/oplock.h b/fs/smb/server/oplock.h index 9a56eaadd0dd..921e3199e4df 100644 --- a/fs/smb/server/oplock.h +++ b/fs/smb/server/oplock.h @@ -69,8 +69,9 @@ struct oplock_info { struct lease *o_lease; struct list_head op_entry; struct list_head lease_entry; - wait_queue_head_t oplock_q; /* Other server threads */ - wait_queue_head_t oplock_brk; /* oplock breaking wait */ + wait_queue_head_t oplock_q; /* Other server threads */ + wait_queue_head_t oplock_brk; /* oplock breaking wait */ + struct rcu_head rcu; }; struct lease_break_info { diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 95901a78951c..9c44e71e3c3b 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -4,6 +4,7 @@ * Copyright (C) 2018 Samsung Electronics Co., Ltd. */ +#include <crypto/utils.h> #include <linux/inetdevice.h> #include <net/addrconf.h> #include <linux/syscalls.h> @@ -125,6 +126,8 @@ int smb2_get_ksmbd_tcon(struct ksmbd_work *work) pr_err("The first operation in the compound does not have tcon\n"); return -EINVAL; } + if (work->tcon->t_state != TREE_CONNECTED) + return -ENOENT; if (tree_id != UINT_MAX && work->tcon->id != tree_id) { pr_err("tree id(%u) is different with id(%u) in first operation\n", tree_id, work->tcon->id); @@ -1947,6 +1950,7 @@ out_err: } } smb2_set_err_rsp(work); + conn->binding = false; } else { unsigned int iov_len; @@ -2827,7 +2831,11 @@ static int parse_durable_handle_context(struct ksmbd_work *work, goto out; } - dh_info->fp->conn = conn; + if (dh_info->fp->conn) { + ksmbd_put_durable_fd(dh_info->fp); + err = -EBADF; + goto out; + } dh_info->reconnected = true; goto out; } @@ -3011,13 +3019,14 @@ int smb2_open(struct ksmbd_work *work) goto err_out2; } + fp = dh_info.fp; + if (ksmbd_override_fsids(work)) { rc = -ENOMEM; ksmbd_put_durable_fd(dh_info.fp); goto err_out2; } - fp = dh_info.fp; file_info = FILE_OPENED; rc = ksmbd_vfs_getattr(&fp->filp->f_path, &stat); @@ -3615,10 +3624,8 @@ int smb2_open(struct ksmbd_work *work) reconnected_fp: rsp->StructureSize = cpu_to_le16(89); - rcu_read_lock(); - opinfo = rcu_dereference(fp->f_opinfo); + opinfo = opinfo_get(fp); rsp->OplockLevel = opinfo != NULL ? opinfo->level : 0; - rcu_read_unlock(); rsp->Flags = 0; rsp->CreateAction = cpu_to_le32(file_info); rsp->CreationTime = cpu_to_le64(fp->create_time); @@ -3659,6 +3666,7 @@ reconnected_fp: next_ptr = &lease_ccontext->Next; next_off = conn->vals->create_lease_size; } + opinfo_put(opinfo); if (maximal_access_ctxt) { struct create_context *mxac_ccontext; @@ -5451,7 +5459,6 @@ static int smb2_get_info_filesystem(struct ksmbd_work *work, struct smb2_query_info_req *req, struct smb2_query_info_rsp *rsp) { - struct ksmbd_session *sess = work->sess; struct ksmbd_conn *conn = work->conn; struct ksmbd_share_config *share = work->tcon->share_conf; int fsinfoclass = 0; @@ -5588,10 +5595,11 @@ static int smb2_get_info_filesystem(struct ksmbd_work *work, info = (struct object_id_info *)(rsp->Buffer); - if (!user_guest(sess->user)) - memcpy(info->objid, user_passkey(sess->user), 16); + if (path.mnt->mnt_sb->s_uuid_len == 16) + memcpy(info->objid, path.mnt->mnt_sb->s_uuid.b, + path.mnt->mnt_sb->s_uuid_len); else - memset(info->objid, 0, 16); + memcpy(info->objid, &stfs.f_fsid, sizeof(stfs.f_fsid)); info->extended_info.magic = cpu_to_le32(EXTENDED_INFO_MAGIC); info->extended_info.version = cpu_to_le32(1); @@ -8880,7 +8888,7 @@ int smb2_check_sign_req(struct ksmbd_work *work) ksmbd_sign_smb2_pdu(work->conn, work->sess->sess_key, iov, 1, signature); - if (memcmp(signature, signature_req, SMB2_SIGNATURE_SIZE)) { + if (crypto_memneq(signature, signature_req, SMB2_SIGNATURE_SIZE)) { pr_err("bad smb2 signature\n"); return 0; } @@ -8968,7 +8976,7 @@ int smb3_check_sign_req(struct ksmbd_work *work) if (ksmbd_sign_smb3_pdu(conn, signing_key, iov, 1, signature)) return 0; - if (memcmp(signature, signature_req, SMB2_SIGNATURE_SIZE)) { + if (crypto_memneq(signature, signature_req, SMB2_SIGNATURE_SIZE)) { pr_err("bad smb2 signature\n"); return 0; } diff --git a/fs/smb/server/smb2pdu.h b/fs/smb/server/smb2pdu.h index 257c6d26df26..8b6eafb70dca 100644 --- a/fs/smb/server/smb2pdu.h +++ b/fs/smb/server/smb2pdu.h @@ -83,7 +83,10 @@ struct create_durable_rsp { } Data; } __packed; -/* equivalent of the contents of SMB3.1.1 POSIX open context response */ +/* + * See POSIX-SMB2 2.2.14.2.16 + * Link: https://gitlab.com/samba-team/smb3-posix-spec/-/blob/master/smb3_posix_extensions.md + */ struct create_posix_rsp { struct create_context_hdr ccontext; __u8 Name[16]; diff --git a/fs/smb/server/transport_rdma.c b/fs/smb/server/transport_rdma.c index 7c53b78b818e..188572491d53 100644 --- a/fs/smb/server/transport_rdma.c +++ b/fs/smb/server/transport_rdma.c @@ -2540,9 +2540,9 @@ static int smb_direct_prepare(struct ksmbd_transport *t) goto put; req = (struct smbdirect_negotiate_req *)recvmsg->packet; - sp->max_recv_size = min_t(int, sp->max_recv_size, + sp->max_recv_size = min_t(u32, sp->max_recv_size, le32_to_cpu(req->preferred_send_size)); - sp->max_send_size = min_t(int, sp->max_send_size, + sp->max_send_size = min_t(u32, sp->max_send_size, le32_to_cpu(req->max_receive_size)); sp->max_fragmented_send_size = le32_to_cpu(req->max_fragmented_size); diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index ff4ea412d900..168f2dd7e200 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -87,11 +87,7 @@ static int proc_show_files(struct seq_file *m, void *v) rcu_read_lock(); opinfo = rcu_dereference(fp->f_opinfo); - rcu_read_unlock(); - - if (!opinfo) { - seq_printf(m, " %-15s", " "); - } else { + if (opinfo) { const struct ksmbd_const_name *const_names; int count; unsigned int level; @@ -105,8 +101,12 @@ static int proc_show_files(struct seq_file *m, void *v) count = ARRAY_SIZE(ksmbd_oplock_const_names); level = opinfo->level; } + rcu_read_unlock(); ksmbd_proc_show_const_name(m, " %-15s", const_names, count, level); + } else { + rcu_read_unlock(); + seq_printf(m, " %-15s", " "); } seq_printf(m, " %#010x %#010x %s\n", diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index 8e958db5f786..67abd4dff222 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -344,6 +344,9 @@ int squashfs_read_metadata(struct super_block *sb, void *buffer, if (unlikely(length < 0)) return -EIO; + if (unlikely(*offset < 0 || *offset >= SQUASHFS_METADATA_SIZE)) + return -EIO; + while (length) { entry = squashfs_cache_get(sb, msblk->block_cache, *block, 0); if (entry->error) { diff --git a/fs/verity/Kconfig b/fs/verity/Kconfig index 76d1c5971b82..b20882963ffb 100644 --- a/fs/verity/Kconfig +++ b/fs/verity/Kconfig @@ -2,6 +2,9 @@ config FS_VERITY bool "FS Verity (read-only file-based authenticity protection)" + # Filesystems cache the Merkle tree at a 64K aligned offset in the + # pagecache. That approach assumes the page size is at most 64K. + depends on PAGE_SHIFT <= 16 select CRYPTO_HASH_INFO select CRYPTO_LIB_SHA256 select CRYPTO_LIB_SHA512 diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index 9c6765cc2d44..bd8fbb40b49e 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -872,6 +872,34 @@ resv_err: return err2; } +void +xfs_growfs_compute_deltas( + struct xfs_mount *mp, + xfs_rfsblock_t nb, + int64_t *deltap, + xfs_agnumber_t *nagcountp) +{ + xfs_rfsblock_t nb_div, nb_mod; + int64_t delta; + xfs_agnumber_t nagcount; + + nb_div = nb; + nb_mod = do_div(nb_div, mp->m_sb.sb_agblocks); + if (nb_mod && nb_mod >= XFS_MIN_AG_BLOCKS) + nb_div++; + else if (nb_mod) + nb = nb_div * mp->m_sb.sb_agblocks; + + if (nb_div > XFS_MAX_AGNUMBER + 1) { + nb_div = XFS_MAX_AGNUMBER + 1; + nb = nb_div * mp->m_sb.sb_agblocks; + } + nagcount = nb_div; + delta = nb - mp->m_sb.sb_dblocks; + *deltap = delta; + *nagcountp = nagcount; +} + /* * Extent the AG indicated by the @id by the length passed in */ diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h index 1f24cfa27321..3cd4790768ff 100644 --- a/fs/xfs/libxfs/xfs_ag.h +++ b/fs/xfs/libxfs/xfs_ag.h @@ -331,6 +331,9 @@ struct aghdr_init_data { int xfs_ag_init_headers(struct xfs_mount *mp, struct aghdr_init_data *id); int xfs_ag_shrink_space(struct xfs_perag *pag, struct xfs_trans **tpp, xfs_extlen_t delta); +void +xfs_growfs_compute_deltas(struct xfs_mount *mp, xfs_rfsblock_t nb, + int64_t *deltap, xfs_agnumber_t *nagcountp); int xfs_ag_extend_space(struct xfs_perag *pag, struct xfs_trans *tp, xfs_extlen_t len); int xfs_ag_get_geometry(struct xfs_perag *pag, struct xfs_ag_geometry *ageo); diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c index 766631f0562e..09d4c17b3e7b 100644 --- a/fs/xfs/libxfs/xfs_da_btree.c +++ b/fs/xfs/libxfs/xfs_da_btree.c @@ -2716,12 +2716,8 @@ xfs_dabuf_map( * larger one that needs to be free by the caller. */ if (nirecs > 1) { - map = kzalloc(nirecs * sizeof(struct xfs_buf_map), - GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL); - if (!map) { - error = -ENOMEM; - goto out_free_irecs; - } + map = kcalloc(nirecs, sizeof(struct xfs_buf_map), + GFP_KERNEL | __GFP_NOLOCKDEP | __GFP_NOFAIL); *mapp = map; } diff --git a/fs/xfs/libxfs/xfs_defer.c b/fs/xfs/libxfs/xfs_defer.c index 472c261163ed..c6909716b041 100644 --- a/fs/xfs/libxfs/xfs_defer.c +++ b/fs/xfs/libxfs/xfs_defer.c @@ -809,7 +809,7 @@ xfs_defer_can_append( /* Paused items cannot absorb more work */ if (dfp->dfp_flags & XFS_DEFER_PAUSED) - return NULL; + return false; /* Already full? */ if (ops->max_items && dfp->dfp_count >= ops->max_items) diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c index a017016e9075..3794e5412eba 100644 --- a/fs/xfs/libxfs/xfs_inode_buf.c +++ b/fs/xfs/libxfs/xfs_inode_buf.c @@ -268,6 +268,10 @@ xfs_inode_from_disk( } if (xfs_is_reflink_inode(ip)) xfs_ifork_init_cow(ip); + if (xfs_is_metadir_inode(ip)) { + XFS_STATS_DEC(ip->i_mount, xs_inodes_active); + XFS_STATS_INC(ip->i_mount, xs_inodes_meta); + } return 0; out_destroy_data_fork: diff --git a/fs/xfs/libxfs/xfs_metafile.c b/fs/xfs/libxfs/xfs_metafile.c index cf239f862212..71f004e9dc64 100644 --- a/fs/xfs/libxfs/xfs_metafile.c +++ b/fs/xfs/libxfs/xfs_metafile.c @@ -61,6 +61,9 @@ xfs_metafile_set_iflag( ip->i_diflags2 |= XFS_DIFLAG2_METADATA; ip->i_metatype = metafile_type; xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + XFS_STATS_DEC(ip->i_mount, xs_inodes_active); + XFS_STATS_INC(ip->i_mount, xs_inodes_meta); } /* Clear the metadata directory inode flag. */ @@ -74,6 +77,8 @@ xfs_metafile_clear_iflag( ip->i_diflags2 &= ~XFS_DIFLAG2_METADATA; xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + XFS_STATS_INC(ip->i_mount, xs_inodes_active); + XFS_STATS_DEC(ip->i_mount, xs_inodes_meta); } /* diff --git a/fs/xfs/libxfs/xfs_ondisk.h b/fs/xfs/libxfs/xfs_ondisk.h index 2e9715cc1641..23cde1248f01 100644 --- a/fs/xfs/libxfs/xfs_ondisk.h +++ b/fs/xfs/libxfs/xfs_ondisk.h @@ -73,7 +73,7 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_free_hdr, 64); XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_leaf, 64); XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_leaf_hdr, 64); - XFS_CHECK_STRUCT_SIZE(struct xfs_attr_leaf_entry, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_attr_leaf_entry, 8); XFS_CHECK_STRUCT_SIZE(struct xfs_attr_leaf_hdr, 32); XFS_CHECK_STRUCT_SIZE(struct xfs_attr_leaf_map, 4); XFS_CHECK_STRUCT_SIZE(struct xfs_attr_leaf_name_local, 4); @@ -116,7 +116,7 @@ xfs_check_ondisk_structs(void) XFS_CHECK_STRUCT_SIZE(struct xfs_da_intnode, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_da_node_entry, 8); XFS_CHECK_STRUCT_SIZE(struct xfs_da_node_hdr, 16); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_free, 4); + XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_free, 4); XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_hdr, 16); XFS_CHECK_OFFSET(struct xfs_dir2_data_unused, freetag, 0); XFS_CHECK_OFFSET(struct xfs_dir2_data_unused, length, 2); @@ -136,16 +136,7 @@ xfs_check_ondisk_structs(void) /* ondisk dir/attr structures from xfs/122 */ XFS_CHECK_STRUCT_SIZE(struct xfs_attr_sf_entry, 3); XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_free, 4); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_hdr, 16); XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_data_unused, 6); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_free, 16); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_free_hdr, 16); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_leaf, 16); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_leaf_entry, 8); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_leaf_hdr, 16); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_leaf_tail, 4); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_sf_entry, 3); - XFS_CHECK_STRUCT_SIZE(struct xfs_dir2_sf_hdr, 10); /* log structures */ XFS_CHECK_STRUCT_SIZE(struct xfs_buf_log_format, 88); @@ -217,11 +208,6 @@ xfs_check_ondisk_structs(void) XFS_CHECK_OFFSET(struct xfs_dir3_free, hdr.hdr.magic, 0); XFS_CHECK_OFFSET(struct xfs_attr3_leafblock, hdr.info.hdr, 0); - XFS_CHECK_STRUCT_SIZE(struct xfs_bulkstat, 192); - XFS_CHECK_STRUCT_SIZE(struct xfs_inumbers, 24); - XFS_CHECK_STRUCT_SIZE(struct xfs_bulkstat_req, 64); - XFS_CHECK_STRUCT_SIZE(struct xfs_inumbers_req, 64); - /* * Make sure the incore inode timestamp range corresponds to hand * converted values based on the ondisk format specification. @@ -301,6 +287,40 @@ xfs_check_ondisk_structs(void) XFS_CHECK_SB_OFFSET(sb_pad, 281); XFS_CHECK_SB_OFFSET(sb_rtstart, 288); XFS_CHECK_SB_OFFSET(sb_rtreserved, 296); + + /* + * ioctl UABI + * + * Due to different padding/alignment requirements across + * different architectures, some structures are ommited from + * the size checks. In addition, structures with architecture + * dependent size fields are also ommited (e.g. __kernel_long_t). + */ + XFS_CHECK_STRUCT_SIZE(struct xfs_bulkstat, 192); + XFS_CHECK_STRUCT_SIZE(struct xfs_inumbers, 24); + XFS_CHECK_STRUCT_SIZE(struct xfs_bulkstat_req, 64); + XFS_CHECK_STRUCT_SIZE(struct xfs_inumbers_req, 64); + XFS_CHECK_STRUCT_SIZE(struct dioattr, 12); + XFS_CHECK_STRUCT_SIZE(struct getbmap, 32); + XFS_CHECK_STRUCT_SIZE(struct getbmapx, 48); + XFS_CHECK_STRUCT_SIZE(struct xfs_attrlist_cursor, 16); + XFS_CHECK_STRUCT_SIZE(struct xfs_attrlist, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_attrlist, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_attrlist_ent, 4); + XFS_CHECK_STRUCT_SIZE(struct xfs_ag_geometry, 128); + XFS_CHECK_STRUCT_SIZE(struct xfs_rtgroup_geometry, 128); + XFS_CHECK_STRUCT_SIZE(struct xfs_error_injection, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_fsop_geom, 256); + XFS_CHECK_STRUCT_SIZE(struct xfs_fsop_geom_v4, 112); + XFS_CHECK_STRUCT_SIZE(struct xfs_fsop_counts, 32); + XFS_CHECK_STRUCT_SIZE(struct xfs_fsop_resblks, 16); + XFS_CHECK_STRUCT_SIZE(struct xfs_growfs_log, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_bulk_ireq, 64); + XFS_CHECK_STRUCT_SIZE(struct xfs_fs_eofblocks, 128); + XFS_CHECK_STRUCT_SIZE(struct xfs_fsid, 8); + XFS_CHECK_STRUCT_SIZE(struct xfs_scrub_metadata, 64); + XFS_CHECK_STRUCT_SIZE(struct xfs_scrub_vec, 16); + XFS_CHECK_STRUCT_SIZE(struct xfs_scrub_vec_head, 40); } #endif /* __XFS_ONDISK_H */ diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index 38d16fe1f6d8..47322adb7690 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -1347,6 +1347,9 @@ xfs_log_sb( * feature was introduced. This counter can go negative due to the way * we handle nearly-lockless reservations, so we must use the _positive * variant here to avoid writing out nonsense frextents. + * + * RT groups are only supported on v5 file systems, which always + * have lazy SB counters. */ if (xfs_has_rtgroups(mp) && !xfs_has_zoned(mp)) { mp->m_sb.sb_frextents = diff --git a/fs/xfs/scrub/dir_repair.c b/fs/xfs/scrub/dir_repair.c index 9dc55c918c78..23b80c54aa60 100644 --- a/fs/xfs/scrub/dir_repair.c +++ b/fs/xfs/scrub/dir_repair.c @@ -177,7 +177,7 @@ xrep_dir_teardown( rd->dir_names = NULL; if (rd->dir_entries) xfarray_destroy(rd->dir_entries); - rd->dir_names = NULL; + rd->dir_entries = NULL; } /* Set up for a directory repair. */ diff --git a/fs/xfs/scrub/orphanage.c b/fs/xfs/scrub/orphanage.c index 52a108f6d5f4..33c6db6b4498 100644 --- a/fs/xfs/scrub/orphanage.c +++ b/fs/xfs/scrub/orphanage.c @@ -442,6 +442,11 @@ xrep_adoption_check_dcache( return 0; d_child = try_lookup_noperm(&qname, d_orphanage); + if (IS_ERR(d_child)) { + dput(d_orphanage); + return PTR_ERR(d_child); + } + if (d_child) { trace_xrep_adoption_check_child(sc->mp, d_child); @@ -479,7 +484,7 @@ xrep_adoption_zap_dcache( return; d_child = try_lookup_noperm(&qname, d_orphanage); - while (d_child != NULL) { + while (!IS_ERR_OR_NULL(d_child)) { trace_xrep_adoption_invalidate_child(sc->mp, d_child); ASSERT(d_is_negative(d_child)); diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c index e8775f254c89..b237a25d6045 100644 --- a/fs/xfs/xfs_bmap_item.c +++ b/fs/xfs/xfs_bmap_item.c @@ -245,7 +245,7 @@ xfs_bmap_update_diff_items( struct xfs_bmap_intent *ba = bi_entry(a); struct xfs_bmap_intent *bb = bi_entry(b); - return ba->bi_owner->i_ino - bb->bi_owner->i_ino; + return cmp_int(ba->bi_owner->i_ino, bb->bi_owner->i_ino); } /* Log bmap updates in the intent item. */ diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 2b208e2c5264..69e9bc588c8b 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -1439,9 +1439,15 @@ xfs_qm_dqflush( return 0; out_abort: + /* + * Shut down the log before removing the dquot item from the AIL. + * Otherwise, the log tail may advance past this item's LSN while + * log writes are still in progress, making these unflushed changes + * unrecoverable on the next mount. + */ + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); dqp->q_flags &= ~XFS_DQFLAG_DIRTY; xfs_trans_ail_delete(lip, 0); - xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); xfs_dqfunlock(dqp); return error; } diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 17255c41786b..8d64d904d73c 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -95,18 +95,17 @@ xfs_growfs_data_private( struct xfs_growfs_data *in) /* growfs data input struct */ { xfs_agnumber_t oagcount = mp->m_sb.sb_agcount; + xfs_rfsblock_t nb = in->newblocks; struct xfs_buf *bp; int error; xfs_agnumber_t nagcount; xfs_agnumber_t nagimax = 0; - xfs_rfsblock_t nb, nb_div, nb_mod; int64_t delta; bool lastag_extended = false; struct xfs_trans *tp; struct aghdr_init_data id = {}; struct xfs_perag *last_pag; - nb = in->newblocks; error = xfs_sb_validate_fsb_count(&mp->m_sb, nb); if (error) return error; @@ -125,20 +124,8 @@ xfs_growfs_data_private( mp->m_sb.sb_rextsize); if (error) return error; + xfs_growfs_compute_deltas(mp, nb, &delta, &nagcount); - nb_div = nb; - nb_mod = do_div(nb_div, mp->m_sb.sb_agblocks); - if (nb_mod && nb_mod >= XFS_MIN_AG_BLOCKS) - nb_div++; - else if (nb_mod) - nb = nb_div * mp->m_sb.sb_agblocks; - - if (nb_div > XFS_MAX_AGNUMBER + 1) { - nb_div = XFS_MAX_AGNUMBER + 1; - nb = nb_div * mp->m_sb.sb_agblocks; - } - nagcount = nb_div; - delta = nb - mp->m_sb.sb_dblocks; /* * Reject filesystems with a single AG because they are not * supported, and reject a shrink operation that would cause a diff --git a/fs/xfs/xfs_health.c b/fs/xfs/xfs_health.c index 169123772cb3..239b843e83d4 100644 --- a/fs/xfs/xfs_health.c +++ b/fs/xfs/xfs_health.c @@ -314,6 +314,22 @@ xfs_rgno_mark_sick( xfs_rtgroup_put(rtg); } +static inline void xfs_inode_report_fserror(struct xfs_inode *ip) +{ + /* + * Do not report inodes being constructed or freed, or metadata inodes, + * to fsnotify. + */ + if (xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIM) || + xfs_is_internal_inode(ip)) { + fserror_report_metadata(ip->i_mount->m_super, -EFSCORRUPTED, + GFP_NOFS); + return; + } + + fserror_report_file_metadata(VFS_I(ip), -EFSCORRUPTED, GFP_NOFS); +} + /* Mark the unhealthy parts of an inode. */ void xfs_inode_mark_sick( @@ -339,7 +355,7 @@ xfs_inode_mark_sick( inode_state_clear(VFS_I(ip), I_DONTCACHE); spin_unlock(&VFS_I(ip)->i_lock); - fserror_report_file_metadata(VFS_I(ip), -EFSCORRUPTED, GFP_NOFS); + xfs_inode_report_fserror(ip); if (mask) xfs_healthmon_report_inode(ip, XFS_HEALTHMON_SICK, old_mask, mask); @@ -371,7 +387,7 @@ xfs_inode_mark_corrupt( inode_state_clear(VFS_I(ip), I_DONTCACHE); spin_unlock(&VFS_I(ip)->i_lock); - fserror_report_file_metadata(VFS_I(ip), -EFSCORRUPTED, GFP_NOFS); + xfs_inode_report_fserror(ip); if (mask) xfs_healthmon_report_inode(ip, XFS_HEALTHMON_CORRUPT, old_mask, mask); diff --git a/fs/xfs/xfs_healthmon.c b/fs/xfs/xfs_healthmon.c index e37c18cec372..26c325d34bd1 100644 --- a/fs/xfs/xfs_healthmon.c +++ b/fs/xfs/xfs_healthmon.c @@ -69,7 +69,7 @@ xfs_healthmon_get( struct xfs_healthmon *hm; rcu_read_lock(); - hm = mp->m_healthmon; + hm = rcu_dereference(mp->m_healthmon); if (hm && !refcount_inc_not_zero(&hm->ref)) hm = NULL; rcu_read_unlock(); @@ -110,13 +110,13 @@ xfs_healthmon_attach( struct xfs_healthmon *hm) { spin_lock(&xfs_healthmon_lock); - if (mp->m_healthmon != NULL) { + if (rcu_access_pointer(mp->m_healthmon) != NULL) { spin_unlock(&xfs_healthmon_lock); return -EEXIST; } refcount_inc(&hm->ref); - mp->m_healthmon = hm; + rcu_assign_pointer(mp->m_healthmon, hm); hm->mount_cookie = (uintptr_t)mp->m_super; spin_unlock(&xfs_healthmon_lock); @@ -128,16 +128,29 @@ STATIC void xfs_healthmon_detach( struct xfs_healthmon *hm) { + struct xfs_mount *mp; + spin_lock(&xfs_healthmon_lock); if (hm->mount_cookie == DETACHED_MOUNT_COOKIE) { spin_unlock(&xfs_healthmon_lock); return; } - XFS_M((struct super_block *)hm->mount_cookie)->m_healthmon = NULL; + mp = XFS_M((struct super_block *)hm->mount_cookie); + rcu_assign_pointer(mp->m_healthmon, NULL); hm->mount_cookie = DETACHED_MOUNT_COOKIE; spin_unlock(&xfs_healthmon_lock); + /* + * Wake up any readers that might remain. This can happen if unmount + * races with the healthmon fd owner entering ->read_iter, having + * already emptied the event queue. + * + * In the ->release case there shouldn't be any readers because the + * only users of the waiter are read and poll. + */ + wake_up_all(&hm->wait); + trace_xfs_healthmon_detach(hm); xfs_healthmon_put(hm); } @@ -1024,13 +1037,6 @@ xfs_healthmon_release( * process can create another health monitor file. */ xfs_healthmon_detach(hm); - - /* - * Wake up any readers that might be left. There shouldn't be any - * because the only users of the waiter are read and poll. - */ - wake_up_all(&hm->wait); - xfs_healthmon_put(hm); return 0; } diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index dbaab4ae709f..2040a9292ee6 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -106,7 +106,7 @@ xfs_inode_alloc( mapping_set_folio_min_order(VFS_I(ip)->i_mapping, M_IGEO(mp)->min_folio_order); - XFS_STATS_INC(mp, vn_active); + XFS_STATS_INC(mp, xs_inodes_active); ASSERT(atomic_read(&ip->i_pincount) == 0); ASSERT(ip->i_ino == 0); @@ -159,7 +159,6 @@ xfs_inode_free_callback( ASSERT(!test_bit(XFS_LI_IN_AIL, &ip->i_itemp->ili_item.li_flags)); xfs_inode_item_destroy(ip); - ip->i_itemp = NULL; } kmem_cache_free(xfs_inode_cache, ip); @@ -172,7 +171,10 @@ __xfs_inode_free( /* asserts to verify all state is correct here */ ASSERT(atomic_read(&ip->i_pincount) == 0); ASSERT(!ip->i_itemp || list_empty(&ip->i_itemp->ili_item.li_bio_list)); - XFS_STATS_DEC(ip->i_mount, vn_active); + if (xfs_is_metadir_inode(ip)) + XFS_STATS_DEC(ip->i_mount, xs_inodes_meta); + else + XFS_STATS_DEC(ip->i_mount, xs_inodes_active); call_rcu(&VFS_I(ip)->i_rcu, xfs_inode_free_callback); } @@ -636,6 +638,14 @@ xfs_iget_cache_miss( if (!ip) return -ENOMEM; + /* + * Set XFS_INEW as early as possible so that the health code won't pass + * the inode to the fserror code if the ondisk inode cannot be loaded. + * We're going to free the xfs_inode immediately if that happens, which + * would lead to UAF problems. + */ + xfs_iflags_set(ip, XFS_INEW); + error = xfs_imap(pag, tp, ip->i_ino, &ip->i_imap, flags); if (error) goto out_destroy; @@ -713,7 +723,6 @@ xfs_iget_cache_miss( ip->i_udquot = NULL; ip->i_gdquot = NULL; ip->i_pdquot = NULL; - xfs_iflags_set(ip, XFS_INEW); /* insert the new inode */ spin_lock(&pag->pag_ici_lock); @@ -2234,7 +2243,7 @@ xfs_inode_mark_reclaimable( struct xfs_mount *mp = ip->i_mount; bool need_inactive; - XFS_STATS_INC(mp, vn_reclaim); + XFS_STATS_INC(mp, xs_inode_mark_reclaimable); /* * We should never get here with any of the reclaim flags already set. diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index b96f262ba139..f807f8f4f705 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1357,6 +1357,8 @@ xlog_alloc_log( if (xfs_has_logv2(mp) && mp->m_sb.sb_logsunit > 1) log->l_iclog_roundoff = mp->m_sb.sb_logsunit; + else if (mp->m_sb.sb_logsectsize > 0) + log->l_iclog_roundoff = mp->m_sb.sb_logsectsize; else log->l_iclog_roundoff = BBSIZE; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 61c71128d171..ddd4028be8d6 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -345,7 +345,7 @@ typedef struct xfs_mount { struct xfs_hooks m_dir_update_hooks; /* Private data referring to a health monitor object. */ - struct xfs_healthmon *m_healthmon; + struct xfs_healthmon __rcu *m_healthmon; } xfs_mount_t; #define M_IGEO(mp) (&(mp)->m_ino_geo) diff --git a/fs/xfs/xfs_notify_failure.c b/fs/xfs/xfs_notify_failure.c index 6be19fa1ebe2..64c8afb935c2 100644 --- a/fs/xfs/xfs_notify_failure.c +++ b/fs/xfs/xfs_notify_failure.c @@ -304,7 +304,7 @@ xfs_dax_notify_dev_failure( error = xfs_alloc_read_agf(pag, tp, 0, &agf_bp); if (error) { - xfs_perag_put(pag); + xfs_perag_rele(pag); break; } @@ -340,7 +340,7 @@ xfs_dax_notify_dev_failure( if (rtg) xfs_rtgroup_unlock(rtg, XFS_RTGLOCK_RMAP); if (error) { - xfs_group_put(xg); + xfs_group_rele(xg); break; } } diff --git a/fs/xfs/xfs_platform.h b/fs/xfs/xfs_platform.h index 1e59bf94d1f2..59a33c60e0ca 100644 --- a/fs/xfs/xfs_platform.h +++ b/fs/xfs/xfs_platform.h @@ -235,6 +235,10 @@ int xfs_rw_bdev(struct block_device *bdev, sector_t sector, unsigned int count, #ifdef XFS_WARN +/* + * Please note that this ASSERT doesn't kill the kernel. It will if the kernel + * has panic_on_warn set. + */ #define ASSERT(expr) \ (likely(expr) ? (void)0 : asswarn(NULL, #expr, __FILE__, __LINE__)) @@ -245,6 +249,11 @@ int xfs_rw_bdev(struct block_device *bdev, sector_t sector, unsigned int count, #endif /* XFS_WARN */ #endif /* DEBUG */ +/* + * Use this to catch metadata corruptions that are not caught by block or + * structure verifiers. The reason is that the verifiers check corruptions only + * within the scope of the object being verified. + */ #define XFS_IS_CORRUPT(mp, expr) \ (unlikely(expr) ? xfs_corruption_error(#expr, XFS_ERRLEVEL_LOW, (mp), \ NULL, 0, __FILE__, __LINE__, \ diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 90a94a5b6f7e..153f3c378f9f 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -112,6 +112,10 @@ xfs_rtcopy_summary( error = xfs_rtget_summary(oargs, log, bbno, &sum); if (error) goto out; + if (XFS_IS_CORRUPT(oargs->mp, sum < 0)) { + error = -EFSCORRUPTED; + goto out; + } if (sum == 0) continue; error = xfs_rtmodify_summary(oargs, log, bbno, -sum); @@ -120,7 +124,6 @@ xfs_rtcopy_summary( error = xfs_rtmodify_summary(nargs, log, bbno, sum); if (error) goto out; - ASSERT(sum > 0); } } error = 0; @@ -1047,6 +1050,15 @@ xfs_growfs_rt_bmblock( */ xfs_trans_resv_calc(mp, &mp->m_resv); + /* + * Sync sb counters now to reflect the updated values. Lazy counters are + * not always updated and in order to avoid inconsistencies between + * frextents and rtextents, it is better to sync the counters. + */ + + if (xfs_has_lazysbcount(mp)) + xfs_log_sb(args.tp); + error = xfs_trans_commit(args.tp); if (error) goto out_free; @@ -1079,17 +1091,27 @@ xfs_last_rtgroup_extents( } /* - * Calculate the last rbmblock currently used. + * This will return the bitmap block number (indexed at 0) that will be + * extended/modified. There are 2 cases here: + * 1. The size of the rtg is such that it is a multiple of + * xfs_rtbitmap_rtx_per_rbmblock() i.e, an integral number of bitmap blocks + * are completely filled up. In this case, we should return + * 1 + (the last used bitmap block number). + * 2. The size of the rtg is not an multiple of xfs_rtbitmap_rtx_per_rbmblock(). + * Here we will return the block number of last used block number. In this + * case, we will modify the last used bitmap block to extend the size of the + * rtgroup. * * This also deals with the case where there were no rtextents before. */ static xfs_fileoff_t -xfs_last_rt_bmblock( +xfs_last_rt_bmblock_to_extend( struct xfs_rtgroup *rtg) { struct xfs_mount *mp = rtg_mount(rtg); xfs_rgnumber_t rgno = rtg_rgno(rtg); xfs_fileoff_t bmbno = 0; + unsigned int mod = 0; ASSERT(!mp->m_sb.sb_rgcount || rgno >= mp->m_sb.sb_rgcount - 1); @@ -1097,9 +1119,16 @@ xfs_last_rt_bmblock( xfs_rtxnum_t nrext = xfs_last_rtgroup_extents(mp); /* Also fill up the previous block if not entirely full. */ - bmbno = xfs_rtbitmap_blockcount_len(mp, nrext); - if (xfs_rtx_to_rbmword(mp, nrext) != 0) - bmbno--; + /* We are doing a -1 to convert it to a 0 based index */ + bmbno = xfs_rtbitmap_blockcount_len(mp, nrext) - 1; + div_u64_rem(nrext, xfs_rtbitmap_rtx_per_rbmblock(mp), &mod); + /* + * mod = 0 means that all the current blocks are full. So + * return the next block number to be used for the rtgroup + * growth. + */ + if (mod == 0) + bmbno++; } return bmbno; @@ -1204,7 +1233,8 @@ xfs_growfs_rtg( goto out_rele; } - for (bmbno = xfs_last_rt_bmblock(rtg); bmbno < bmblocks; bmbno++) { + for (bmbno = xfs_last_rt_bmblock_to_extend(rtg); bmbno < bmblocks; + bmbno++) { error = xfs_growfs_rt_bmblock(rtg, nrblocks, rextsize, bmbno); if (error) goto out_error; diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c index 017db0361cd8..c13d600732c9 100644 --- a/fs/xfs/xfs_stats.c +++ b/fs/xfs/xfs_stats.c @@ -42,7 +42,7 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf) { "xstrat", xfsstats_offset(xs_write_calls) }, { "rw", xfsstats_offset(xs_attr_get) }, { "attr", xfsstats_offset(xs_iflush_count)}, - { "icluster", xfsstats_offset(vn_active) }, + { "icluster", xfsstats_offset(xs_inodes_active) }, { "vnodes", xfsstats_offset(xb_get) }, { "buf", xfsstats_offset(xs_abtb_2) }, { "abtb2", xfsstats_offset(xs_abtc_2) }, @@ -59,7 +59,8 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf) { "rtrefcntbt", xfsstats_offset(xs_qm_dqreclaims)}, /* we print both series of quota information together */ { "qm", xfsstats_offset(xs_gc_read_calls)}, - { "zoned", xfsstats_offset(__pad1)}, + { "zoned", xfsstats_offset(xs_inodes_meta)}, + { "metafile", xfsstats_offset(xs_xstrat_bytes)}, }; /* Loop over all stats groups */ @@ -99,16 +100,20 @@ int xfs_stats_format(struct xfsstats __percpu *stats, char *buf) void xfs_stats_clearall(struct xfsstats __percpu *stats) { + uint32_t xs_inodes_active, xs_inodes_meta; int c; - uint32_t vn_active; xfs_notice(NULL, "Clearing xfsstats"); for_each_possible_cpu(c) { preempt_disable(); - /* save vn_active, it's a universal truth! */ - vn_active = per_cpu_ptr(stats, c)->s.vn_active; + /* + * Save the active / meta inode counters, as they are stateful. + */ + xs_inodes_active = per_cpu_ptr(stats, c)->s.xs_inodes_active; + xs_inodes_meta = per_cpu_ptr(stats, c)->s.xs_inodes_meta; memset(per_cpu_ptr(stats, c), 0, sizeof(*stats)); - per_cpu_ptr(stats, c)->s.vn_active = vn_active; + per_cpu_ptr(stats, c)->s.xs_inodes_active = xs_inodes_active; + per_cpu_ptr(stats, c)->s.xs_inodes_meta = xs_inodes_meta; preempt_enable(); } } diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h index 153d2381d0a8..57c32b86c358 100644 --- a/fs/xfs/xfs_stats.h +++ b/fs/xfs/xfs_stats.h @@ -100,14 +100,14 @@ struct __xfsstats { uint32_t xs_iflush_count; uint32_t xs_icluster_flushcnt; uint32_t xs_icluster_flushinode; - uint32_t vn_active; /* # vnodes not on free lists */ - uint32_t vn_alloc; /* # times vn_alloc called */ - uint32_t vn_get; /* # times vn_get called */ - uint32_t vn_hold; /* # times vn_hold called */ - uint32_t vn_rele; /* # times vn_rele called */ - uint32_t vn_reclaim; /* # times vn_reclaim called */ - uint32_t vn_remove; /* # times vn_remove called */ - uint32_t vn_free; /* # times vn_free called */ + uint32_t xs_inodes_active; + uint32_t __unused_vn_alloc; + uint32_t __unused_vn_get; + uint32_t __unused_vn_hold; + uint32_t xs_inode_destroy; + uint32_t xs_inode_destroy2; /* same as xs_inode_destroy */ + uint32_t xs_inode_mark_reclaimable; + uint32_t __unused_vn_free; uint32_t xb_get; uint32_t xb_create; uint32_t xb_get_locked; @@ -142,7 +142,8 @@ struct __xfsstats { uint32_t xs_gc_read_calls; uint32_t xs_gc_write_calls; uint32_t xs_gc_zone_reset_calls; - uint32_t __pad1; +/* Metafile counters */ + uint32_t xs_inodes_meta; /* Extra precision counters */ uint64_t xs_xstrat_bytes; uint64_t xs_write_bytes; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index abc45f860a73..f8de44443e81 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -712,8 +712,8 @@ xfs_fs_destroy_inode( trace_xfs_destroy_inode(ip); ASSERT(!rwsem_is_locked(&inode->i_rwsem)); - XFS_STATS_INC(ip->i_mount, vn_rele); - XFS_STATS_INC(ip->i_mount, vn_remove); + XFS_STATS_INC(ip->i_mount, xs_inode_destroy); + XFS_STATS_INC(ip->i_mount, xs_inode_destroy2); xfs_inode_mark_reclaimable(ip); } diff --git a/fs/xfs/xfs_verify_media.c b/fs/xfs/xfs_verify_media.c index 069cd371619d..8bbd4ec567f8 100644 --- a/fs/xfs/xfs_verify_media.c +++ b/fs/xfs/xfs_verify_media.c @@ -122,7 +122,7 @@ xfs_verify_report_losses( error = xfs_alloc_read_agf(pag, tp, 0, &agf_bp); if (error) { - xfs_perag_put(pag); + xfs_perag_rele(pag); break; } @@ -158,7 +158,7 @@ xfs_verify_report_losses( if (rtg) xfs_rtgroup_unlock(rtg, XFS_RTGLOCK_RMAP); if (error) { - xfs_group_put(xg); + xfs_group_rele(xg); break; } } diff --git a/fs/xfs/xfs_zone_alloc.c b/fs/xfs/xfs_zone_alloc.c index 67e0c8f5800f..e3d19b6dc64a 100644 --- a/fs/xfs/xfs_zone_alloc.c +++ b/fs/xfs/xfs_zone_alloc.c @@ -78,7 +78,7 @@ xfs_zone_account_reclaimable( struct xfs_rtgroup *rtg, uint32_t freed) { - struct xfs_group *xg = &rtg->rtg_group; + struct xfs_group *xg = rtg_group(rtg); struct xfs_mount *mp = rtg_mount(rtg); struct xfs_zone_info *zi = mp->m_zone_info; uint32_t used = rtg_rmap(rtg)->i_used_blocks; @@ -759,7 +759,7 @@ xfs_zone_alloc_blocks( trace_xfs_zone_alloc_blocks(oz, allocated, count_fsb); - *sector = xfs_gbno_to_daddr(&rtg->rtg_group, 0); + *sector = xfs_gbno_to_daddr(rtg_group(rtg), 0); *is_seq = bdev_zone_is_seq(mp->m_rtdev_targp->bt_bdev, *sector); if (!*is_seq) *sector += XFS_FSB_TO_BB(mp, allocated); @@ -1080,7 +1080,7 @@ xfs_init_zone( if (write_pointer == 0) { /* zone is empty */ atomic_inc(&zi->zi_nr_free_zones); - xfs_group_set_mark(&rtg->rtg_group, XFS_RTG_FREE); + xfs_group_set_mark(rtg_group(rtg), XFS_RTG_FREE); iz->available += rtg_blocks(rtg); } else if (write_pointer < rtg_blocks(rtg)) { /* zone is open */ diff --git a/fs/xfs/xfs_zone_gc.c b/fs/xfs/xfs_zone_gc.c index 48c6cf584447..309f70098524 100644 --- a/fs/xfs/xfs_zone_gc.c +++ b/fs/xfs/xfs_zone_gc.c @@ -96,7 +96,6 @@ struct xfs_gc_bio { */ xfs_fsblock_t old_startblock; xfs_daddr_t new_daddr; - struct xfs_zone_scratch *scratch; /* Are we writing to a sequential write required zone? */ bool is_seq; @@ -627,7 +626,7 @@ xfs_zone_gc_alloc_blocks( if (!*count_fsb) return NULL; - *daddr = xfs_gbno_to_daddr(&oz->oz_rtg->rtg_group, 0); + *daddr = xfs_gbno_to_daddr(rtg_group(oz->oz_rtg), 0); *is_seq = bdev_zone_is_seq(mp->m_rtdev_targp->bt_bdev, *daddr); if (!*is_seq) *daddr += XFS_FSB_TO_BB(mp, oz->oz_allocated); @@ -702,7 +701,7 @@ xfs_zone_gc_start_chunk( chunk->data = data; chunk->oz = oz; chunk->victim_rtg = iter->victim_rtg; - atomic_inc(&chunk->victim_rtg->rtg_group.xg_active_ref); + atomic_inc(&rtg_group(chunk->victim_rtg)->xg_active_ref); atomic_inc(&chunk->victim_rtg->rtg_gccount); bio->bi_iter.bi_sector = xfs_rtb_to_daddr(mp, chunk->old_startblock); @@ -779,7 +778,6 @@ xfs_zone_gc_split_write( ihold(VFS_I(chunk->ip)); split_chunk->ip = chunk->ip; split_chunk->is_seq = chunk->is_seq; - split_chunk->scratch = chunk->scratch; split_chunk->offset = chunk->offset; split_chunk->len = split_len; split_chunk->old_startblock = chunk->old_startblock; @@ -788,7 +786,7 @@ xfs_zone_gc_split_write( atomic_inc(&chunk->oz->oz_ref); split_chunk->victim_rtg = chunk->victim_rtg; - atomic_inc(&chunk->victim_rtg->rtg_group.xg_active_ref); + atomic_inc(&rtg_group(chunk->victim_rtg)->xg_active_ref); atomic_inc(&chunk->victim_rtg->rtg_gccount); chunk->offset += split_len; @@ -888,7 +886,7 @@ xfs_zone_gc_finish_reset( goto out; } - xfs_group_set_mark(&rtg->rtg_group, XFS_RTG_FREE); + xfs_group_set_mark(rtg_group(rtg), XFS_RTG_FREE); atomic_inc(&zi->zi_nr_free_zones); xfs_zoned_add_available(mp, rtg_blocks(rtg)); @@ -917,7 +915,7 @@ xfs_submit_zone_reset_bio( XFS_STATS_INC(mp, xs_gc_zone_reset_calls); - bio->bi_iter.bi_sector = xfs_gbno_to_daddr(&rtg->rtg_group, 0); + bio->bi_iter.bi_sector = xfs_gbno_to_daddr(rtg_group(rtg), 0); if (!bdev_zone_is_seq(bio->bi_bdev, bio->bi_iter.bi_sector)) { /* * Also use the bio to drive the state machine when neither |