diff options
Diffstat (limited to 'fs')
-rw-r--r-- | fs/afs/dir.c | 2 | ||||
-rw-r--r-- | fs/afs/fs_probe.c | 11 | ||||
-rw-r--r-- | fs/afs/internal.h | 1 | ||||
-rw-r--r-- | fs/afs/main.c | 3 | ||||
-rw-r--r-- | fs/afs/server.c | 3 | ||||
-rw-r--r-- | fs/block_dev.c | 17 | ||||
-rw-r--r-- | fs/btrfs/block-group.c | 44 | ||||
-rw-r--r-- | fs/btrfs/ctree.h | 2 | ||||
-rw-r--r-- | fs/btrfs/file.c | 46 | ||||
-rw-r--r-- | fs/btrfs/inode.c | 39 | ||||
-rw-r--r-- | fs/btrfs/ioctl.c | 2 | ||||
-rw-r--r-- | fs/btrfs/tree-log.c | 5 | ||||
-rw-r--r-- | fs/erofs/zdata.h | 20 | ||||
-rw-r--r-- | fs/file_table.c | 2 | ||||
-rw-r--r-- | fs/io-wq.c | 108 | ||||
-rw-r--r-- | fs/io-wq.h | 4 | ||||
-rw-r--r-- | fs/io_uring.c | 177 | ||||
-rw-r--r-- | fs/proc/bootconfig.c | 15 | ||||
-rw-r--r-- | fs/proc/kcore.c | 3 |
19 files changed, 328 insertions, 176 deletions
diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 3e3c2bf0a722..96757f3abd74 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -845,7 +845,7 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, * to FS.FetchStatus for op->file[1]. */ op->fetch_status.which = 1; - op->ops = &afs_fetch_status_operation; + op->ops = &afs_lookup_fetch_status_operation; afs_begin_vnode_operation(op); afs_wait_for_operation(op); } diff --git a/fs/afs/fs_probe.c b/fs/afs/fs_probe.c index b34f74b0f319..5d9ef517cf81 100644 --- a/fs/afs/fs_probe.c +++ b/fs/afs/fs_probe.c @@ -314,7 +314,7 @@ void afs_fs_probe_timer(struct timer_list *timer) { struct afs_net *net = container_of(timer, struct afs_net, fs_probe_timer); - if (!queue_work(afs_wq, &net->fs_prober)) + if (!net->live || !queue_work(afs_wq, &net->fs_prober)) afs_dec_servers_outstanding(net); } @@ -458,3 +458,12 @@ dont_wait: return -ETIME; return -EDESTADDRREQ; } + +/* + * Clean up the probing when the namespace is killed off. + */ +void afs_fs_probe_cleanup(struct afs_net *net) +{ + if (del_timer_sync(&net->fs_probe_timer)) + afs_dec_servers_outstanding(net); +} diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 573a5922c3bb..d520535ddb62 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1065,6 +1065,7 @@ extern int afs_wait_for_fs_probes(struct afs_server_list *, unsigned long); extern void afs_probe_fileserver(struct afs_net *, struct afs_server *); extern void afs_fs_probe_dispatcher(struct work_struct *); extern int afs_wait_for_one_fs_probe(struct afs_server *, bool); +extern void afs_fs_probe_cleanup(struct afs_net *); /* * inode.c diff --git a/fs/afs/main.c b/fs/afs/main.c index 9c79c91e8005..31b472f7c734 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -100,6 +100,7 @@ static int __net_init afs_net_init(struct net *net_ns) timer_setup(&net->fs_timer, afs_servers_timer, 0); INIT_WORK(&net->fs_prober, afs_fs_probe_dispatcher); timer_setup(&net->fs_probe_timer, afs_fs_probe_timer, 0); + atomic_set(&net->servers_outstanding, 1); ret = -ENOMEM; sysnames = kzalloc(sizeof(*sysnames), GFP_KERNEL); @@ -130,6 +131,7 @@ static int __net_init afs_net_init(struct net *net_ns) error_open_socket: net->live = false; + afs_fs_probe_cleanup(net); afs_cell_purge(net); afs_purge_servers(net); error_cell_init: @@ -150,6 +152,7 @@ static void __net_exit afs_net_exit(struct net *net_ns) struct afs_net *net = afs_net(net_ns); net->live = false; + afs_fs_probe_cleanup(net); afs_cell_purge(net); afs_purge_servers(net); afs_close_socket(net); diff --git a/fs/afs/server.c b/fs/afs/server.c index 039e3488511c..e82e452e2612 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -605,11 +605,12 @@ void afs_purge_servers(struct afs_net *net) _enter(""); if (del_timer_sync(&net->fs_timer)) - atomic_dec(&net->servers_outstanding); + afs_dec_servers_outstanding(net); afs_queue_server_manager(net); _debug("wait"); + atomic_dec(&net->servers_outstanding); wait_var_event(&net->servers_outstanding, !atomic_read(&net->servers_outstanding)); _leave(""); diff --git a/fs/block_dev.c b/fs/block_dev.c index 47860e589388..0ae656e022fd 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -75,7 +75,7 @@ static void bdev_write_inode(struct block_device *bdev) } /* Kill _all_ buffers and pagecache , dirty or not.. */ -void kill_bdev(struct block_device *bdev) +static void kill_bdev(struct block_device *bdev) { struct address_space *mapping = bdev->bd_inode->i_mapping; @@ -84,8 +84,7 @@ void kill_bdev(struct block_device *bdev) invalidate_bh_lrus(); truncate_inode_pages(mapping, 0); -} -EXPORT_SYMBOL(kill_bdev); +} /* Invalidate clean unused buffers and pagecache. */ void invalidate_bdev(struct block_device *bdev) @@ -1565,10 +1564,8 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) */ if (!for_part) { ret = devcgroup_inode_permission(bdev->bd_inode, perm); - if (ret != 0) { - bdput(bdev); + if (ret != 0) return ret; - } } restart: @@ -1637,8 +1634,10 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) goto out_clear; BUG_ON(for_part); ret = __blkdev_get(whole, mode, 1); - if (ret) + if (ret) { + bdput(whole); goto out_clear; + } bdev->bd_contains = whole; bdev->bd_part = disk_get_part(disk, partno); if (!(disk->flags & GENHD_FL_UP) || @@ -1688,7 +1687,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part) disk_unblock_events(disk); put_disk_and_module(disk); out: - bdput(bdev); return ret; } @@ -1755,6 +1753,9 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) bdput(whole); } + if (res) + bdput(bdev); + return res; } EXPORT_SYMBOL(blkdev_get); diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 176e8a292fd1..c037ef514b64 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -940,7 +940,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; - goto out_put_group; + goto out; } /* @@ -978,7 +978,7 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, ret = btrfs_orphan_add(trans, BTRFS_I(inode)); if (ret) { btrfs_add_delayed_iput(inode); - goto out_put_group; + goto out; } clear_nlink(inode); /* One for the block groups ref */ @@ -1001,13 +1001,13 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, ret = btrfs_search_slot(trans, tree_root, &key, path, -1, 1); if (ret < 0) - goto out_put_group; + goto out; if (ret > 0) btrfs_release_path(path); if (ret == 0) { ret = btrfs_del_item(trans, tree_root, path); if (ret) - goto out_put_group; + goto out; btrfs_release_path(path); } @@ -1016,6 +1016,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, &fs_info->block_group_cache_tree); RB_CLEAR_NODE(&block_group->cache_node); + /* Once for the block groups rbtree */ + btrfs_put_block_group(block_group); + if (fs_info->first_logical_byte == block_group->start) fs_info->first_logical_byte = (u64)-1; spin_unlock(&fs_info->block_group_cache_lock); @@ -1089,6 +1092,25 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, spin_unlock(&block_group->space_info->lock); + /* + * Remove the free space for the block group from the free space tree + * and the block group's item from the extent tree before marking the + * block group as removed. This is to prevent races with tasks that + * freeze and unfreeze a block group, this task and another task + * allocating a new block group - the unfreeze task ends up removing + * the block group's extent map before the task calling this function + * deletes the block group item from the extent tree, allowing for + * another task to attempt to create another block group with the same + * item key (and failing with -EEXIST and a transaction abort). + */ + ret = remove_block_group_free_space(trans, block_group); + if (ret) + goto out; + + ret = remove_block_group_item(trans, path, block_group); + if (ret < 0) + goto out; + mutex_lock(&fs_info->chunk_mutex); spin_lock(&block_group->lock); block_group->removed = 1; @@ -1123,17 +1145,6 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, mutex_unlock(&fs_info->chunk_mutex); - ret = remove_block_group_free_space(trans, block_group); - if (ret) - goto out_put_group; - - /* Once for the block groups rbtree */ - btrfs_put_block_group(block_group); - - ret = remove_block_group_item(trans, path, block_group); - if (ret < 0) - goto out; - if (remove_em) { struct extent_map_tree *em_tree; @@ -1145,10 +1156,9 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, free_extent_map(em); } -out_put_group: +out: /* Once for the lookup reference */ btrfs_put_block_group(block_group); -out: if (remove_rsv) btrfs_delayed_refs_rsv_release(fs_info, 1); btrfs_free_path(path); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 30ce7039bc27..d404cce8ae40 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1009,6 +1009,8 @@ enum { BTRFS_ROOT_DEAD_RELOC_TREE, /* Mark dead root stored on device whose cleanup needs to be resumed */ BTRFS_ROOT_DEAD_TREE, + /* The root has a log tree. Used only for subvolume roots. */ + BTRFS_ROOT_HAS_LOG_TREE, }; /* diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 2c14312b05e8..2520605afc25 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1533,7 +1533,7 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages, } static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, - size_t *write_bytes) + size_t *write_bytes, bool nowait) { struct btrfs_fs_info *fs_info = inode->root->fs_info; struct btrfs_root *root = inode->root; @@ -1541,27 +1541,43 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, u64 num_bytes; int ret; - if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) + if (!nowait && !btrfs_drew_try_write_lock(&root->snapshot_lock)) return -EAGAIN; lockstart = round_down(pos, fs_info->sectorsize); lockend = round_up(pos + *write_bytes, fs_info->sectorsize) - 1; + num_bytes = lockend - lockstart + 1; - btrfs_lock_and_flush_ordered_range(inode, lockstart, - lockend, NULL); + if (nowait) { + struct btrfs_ordered_extent *ordered; + + if (!try_lock_extent(&inode->io_tree, lockstart, lockend)) + return -EAGAIN; + + ordered = btrfs_lookup_ordered_range(inode, lockstart, + num_bytes); + if (ordered) { + btrfs_put_ordered_extent(ordered); + ret = -EAGAIN; + goto out_unlock; + } + } else { + btrfs_lock_and_flush_ordered_range(inode, lockstart, + lockend, NULL); + } - num_bytes = lockend - lockstart + 1; ret = can_nocow_extent(&inode->vfs_inode, lockstart, &num_bytes, NULL, NULL, NULL); if (ret <= 0) { ret = 0; - btrfs_drew_write_unlock(&root->snapshot_lock); + if (!nowait) + btrfs_drew_write_unlock(&root->snapshot_lock); } else { *write_bytes = min_t(size_t, *write_bytes , num_bytes - pos + lockstart); } - +out_unlock: unlock_extent(&inode->io_tree, lockstart, lockend); return ret; @@ -1633,7 +1649,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb, if ((BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)) && check_can_nocow(BTRFS_I(inode), pos, - &write_bytes) > 0) { + &write_bytes, false) > 0) { /* * For nodata cow case, no need to reserve * data space. @@ -1904,13 +1920,25 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, pos = iocb->ki_pos; count = iov_iter_count(from); if (iocb->ki_flags & IOCB_NOWAIT) { + size_t nocow_bytes = count; + /* * We will allocate space in case nodatacow is not set, * so bail */ if (!(BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | BTRFS_INODE_PREALLOC)) || - check_can_nocow(BTRFS_I(inode), pos, &count) <= 0) { + check_can_nocow(BTRFS_I(inode), pos, &nocow_bytes, + true) <= 0) { + inode_unlock(inode); + return -EAGAIN; + } + /* + * There are holes in the range or parts of the range that must + * be COWed (shared extents, RO block groups, etc), so just bail + * out. + */ + if (nocow_bytes < count) { inode_unlock(inode); return -EAGAIN; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index d04c82c88418..18d384f4af54 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -985,6 +985,7 @@ static noinline int cow_file_range(struct inode *inode, u64 num_bytes; unsigned long ram_size; u64 cur_alloc_size = 0; + u64 min_alloc_size; u64 blocksize = fs_info->sectorsize; struct btrfs_key ins; struct extent_map *em; @@ -1035,10 +1036,26 @@ static noinline int cow_file_range(struct inode *inode, btrfs_drop_extent_cache(BTRFS_I(inode), start, start + num_bytes - 1, 0); + /* + * Relocation relies on the relocated extents to have exactly the same + * size as the original extents. Normally writeback for relocation data + * extents follows a NOCOW path because relocation preallocates the + * extents. However, due to an operation such as scrub turning a block + * group to RO mode, it may fallback to COW mode, so we must make sure + * an extent allocated during COW has exactly the requested size and can + * not be split into smaller extents, otherwise relocation breaks and + * fails during the stage where it updates the bytenr of file extent + * items. + */ + if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID) + min_alloc_size = num_bytes; + else + min_alloc_size = fs_info->sectorsize; + while (num_bytes > 0) { cur_alloc_size = num_bytes; ret = btrfs_reserve_extent(root, cur_alloc_size, cur_alloc_size, - fs_info->sectorsize, 0, alloc_hint, + min_alloc_size, 0, alloc_hint, &ins, 1, 1); if (ret < 0) goto out_unlock; @@ -1361,6 +1378,8 @@ static int fallback_to_cow(struct inode *inode, struct page *locked_page, int *page_started, unsigned long *nr_written) { const bool is_space_ino = btrfs_is_free_space_inode(BTRFS_I(inode)); + const bool is_reloc_ino = (BTRFS_I(inode)->root->root_key.objectid == + BTRFS_DATA_RELOC_TREE_OBJECTID); const u64 range_bytes = end + 1 - start; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; u64 range_start = start; @@ -1391,18 +1410,23 @@ static int fallback_to_cow(struct inode *inode, struct page *locked_page, * data space info, which we incremented in the step above. * * If we need to fallback to cow and the inode corresponds to a free - * space cache inode, we must also increment bytes_may_use of the data - * space_info for the same reason. Space caches always get a prealloc + * space cache inode or an inode of the data relocation tree, we must + * also increment bytes_may_use of the data space_info for the same + * reason. Space caches and relocated data extents always get a prealloc * extent for them, however scrub or balance may have set the block - * group that contains that extent to RO mode. + * group that contains that extent to RO mode and therefore force COW + * when starting writeback. */ count = count_range_bits(io_tree, &range_start, end, range_bytes, EXTENT_NORESERVE, 0); - if (count > 0 || is_space_ino) { - const u64 bytes = is_space_ino ? range_bytes : count; + if (count > 0 || is_space_ino || is_reloc_ino) { + u64 bytes = count; struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; struct btrfs_space_info *sinfo = fs_info->data_sinfo; + if (is_space_ino || is_reloc_ino) + bytes = range_bytes; + spin_lock(&sinfo->lock); btrfs_space_info_update_bytes_may_use(fs_info, sinfo, bytes); spin_unlock(&sinfo->lock); @@ -7865,9 +7889,6 @@ static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) dio_data.overwrite = 1; inode_unlock(inode); relock = true; - } else if (iocb->ki_flags & IOCB_NOWAIT) { - ret = -EAGAIN; - goto out; } ret = btrfs_delalloc_reserve_space(inode, &data_reserved, offset, count); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 168deb8ef68a..e8f7c5f00894 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2692,7 +2692,7 @@ out: btrfs_put_root(root); out_free: btrfs_free_path(path); - kzfree(subvol_info); + kfree(subvol_info); return ret; } diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 920cee312f4e..cd5348f352dd 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -169,6 +169,7 @@ static int start_log_trans(struct btrfs_trans_handle *trans, if (ret) goto out; + set_bit(BTRFS_ROOT_HAS_LOG_TREE, &root->state); clear_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); root->log_start_pid = current->pid; } @@ -195,6 +196,9 @@ static int join_running_log_trans(struct btrfs_root *root) { int ret = -ENOENT; + if (!test_bit(BTRFS_ROOT_HAS_LOG_TREE, &root->state)) + return ret; + mutex_lock(&root->log_mutex); if (root->log_root) { ret = 0; @@ -3303,6 +3307,7 @@ int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root) if (root->log_root) { free_log_tree(trans, root->log_root); root->log_root = NULL; + clear_bit(BTRFS_ROOT_HAS_LOG_TREE, &root->state); } return 0; } diff --git a/fs/erofs/zdata.h b/fs/erofs/zdata.h index 7824f5563a55..9b66c28b3ae9 100644 --- a/fs/erofs/zdata.h +++ b/fs/erofs/zdata.h @@ -144,22 +144,22 @@ static inline void z_erofs_onlinepage_init(struct page *page) static inline void z_erofs_onlinepage_fixup(struct page *page, uintptr_t index, bool down) { - unsigned long *p, o, v, id; -repeat: - p = &page_private(page); - o = READ_ONCE(*p); + union z_erofs_onlinepage_converter u = { .v = &page_private(page) }; + int orig, orig_index, val; - id = o >> Z_EROFS_ONLINEPAGE_INDEX_SHIFT; - if (id) { +repeat: + orig = atomic_read(u.o); + orig_index = orig >> Z_EROFS_ONLINEPAGE_INDEX_SHIFT; + if (orig_index) { if (!index) return; - DBG_BUGON(id != index); + DBG_BUGON(orig_index != index); } - v = (index << Z_EROFS_ONLINEPAGE_INDEX_SHIFT) | - ((o & Z_EROFS_ONLINEPAGE_COUNT_MASK) + (unsigned int)down); - if (cmpxchg(p, o, v) != o) + val = (index << Z_EROFS_ONLINEPAGE_INDEX_SHIFT) | + ((orig & Z_EROFS_ONLINEPAGE_COUNT_MASK) + (unsigned int)down); + if (atomic_cmpxchg(u.o, orig, val) != orig) goto repeat; } diff --git a/fs/file_table.c b/fs/file_table.c index 656647f9575a..65603502fed6 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -230,7 +230,7 @@ struct file *alloc_file_pseudo(struct inode *inode, struct vfsmount *mnt, d_set_d_op(path.dentry, &anon_ops); path.mnt = mntget(mnt); d_instantiate(path.dentry, inode); - file = alloc_file(&path, flags, fops); + file = alloc_file(&path, flags | FMODE_NONOTIFY, fops); if (IS_ERR(file)) { ihold(inode); path_put(&path); diff --git a/fs/io-wq.c b/fs/io-wq.c index 0b65a912b036..47c5f3aeb460 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -903,13 +903,15 @@ void io_wq_cancel_all(struct io_wq *wq) struct io_cb_cancel_data { work_cancel_fn *fn; void *data; + int nr_running; + int nr_pending; + bool cancel_all; }; static bool io_wq_worker_cancel(struct io_worker *worker, void *data) { struct io_cb_cancel_data *match = data; unsigned long flags; - bool ret = false; /* * Hold the lock to avoid ->cur_work going out of scope, caller @@ -920,74 +922,90 @@ static bool io_wq_worker_cancel(struct io_worker *worker, void *data) !(worker->cur_work->flags & IO_WQ_WORK_NO_CANCEL) && match->fn(worker->cur_work, match->data)) { send_sig(SIGINT, worker->task, 1); - ret = true; + match->nr_running++; } spin_unlock_irqrestore(&worker->lock, flags); - return ret; + return match->nr_running && !match->cancel_all; } -static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe, - struct io_cb_cancel_data *match) +static void io_wqe_cancel_pending_work(struct io_wqe *wqe, + struct io_cb_cancel_data *match) { struct io_wq_work_node *node, *prev; struct io_wq_work *work; unsigned long flags; - bool found = false; - /* - * First check pending list, if we're lucky we can just remove it - * from there. CANCEL_OK means that the work is returned as-new, - * no completion will be posted for it. - */ +retry: spin_lock_irqsave(&wqe->lock, flags); wq_list_for_each(node, prev, &wqe->work_list) { work = container_of(node, struct io_wq_work, list); + if (!match->fn(work, match->data)) + continue; - if (match->fn(work, match->data)) { - wq_list_del(&wqe->work_list, node, prev); - found = true; - break; - } - } - spin_unlock_irqrestore(&wqe->lock, flags); - - if (found) { + wq_list_del(&wqe->work_list, node, prev); + spin_unlock_irqrestore(&wqe->lock, flags); io_run_cancel(work, wqe); - return IO_WQ_CANCEL_OK; + match->nr_pending++; + if (!match->cancel_all) + return; + + /* not safe to continue after unlock */ + goto retry; } + spin_unlock_irqrestore(&wqe->lock, flags); +} - /* - * Now check if a free (going busy) or busy worker has the work - * currently running. If we find it there, we'll return CANCEL_RUNNING - * as an indication that we attempt to signal cancellation. The - * completion will run normally in this case. - */ +static void io_wqe_cancel_running_work(struct io_wqe *wqe, + struct io_cb_cancel_data *match) +{ rcu_read_lock(); - found = io_wq_for_each_worker(wqe, io_wq_worker_cancel, match); + io_wq_for_each_worker(wqe, io_wq_worker_cancel, match); rcu_read_unlock(); - return found ? IO_WQ_CANCEL_RUNNING : IO_WQ_CANCEL_NOTFOUND; } enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, - void *data) + void *data, bool cancel_all) { struct io_cb_cancel_data match = { - .fn = cancel, - .data = data, + .fn = cancel, + .data = data, + .cancel_all = cancel_all, }; - enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND; int node; + /* + * First check pending list, if we're lucky we can just remove it + * from there. CANCEL_OK means that the work is returned as-new, + * no completion will be posted for it. + */ for_each_node(node) { struct io_wqe *wqe = wq->wqes[node]; - ret = io_wqe_cancel_work(wqe, &match); - if (ret != IO_WQ_CANCEL_NOTFOUND) - break; + io_wqe_cancel_pending_work(wqe, &match); + if (match.nr_pending && !match.cancel_all) + return IO_WQ_CANCEL_OK; } - return ret; + /* + * Now check if a free (going busy) or busy worker has the work + * currently running. If we find it there, we'll return CANCEL_RUNNING + * as an indication that we attempt to signal cancellation. The + * completion will run normally in this case. + */ + for_each_node(node) { + struct io_wqe *wqe = wq->wqes[node]; + + io_wqe_cancel_running_work(wqe, &match); + if (match.nr_running && !match.cancel_all) + return IO_WQ_CANCEL_RUNNING; + } + + if (match.nr_running) + return IO_WQ_CANCEL_RUNNING; + if (match.nr_pending) + return IO_WQ_CANCEL_OK; + return IO_WQ_CANCEL_NOTFOUND; } static bool io_wq_io_cb_cancel_data(struct io_wq_work *work, void *data) @@ -997,21 +1015,7 @@ static bool io_wq_io_cb_cancel_data(struct io_wq_work *work, void *data) enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork) { - return io_wq_cancel_cb(wq, io_wq_io_cb_cancel_data, (void *)cwork); -} - -static bool io_wq_pid_match(struct io_wq_work *work, void *data) -{ - pid_t pid = (pid_t) (unsigned long) data; - - return work->task_pid == pid; -} - -enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid) -{ - void *data = (void *) (unsigned long) pid; - - return io_wq_cancel_cb(wq, io_wq_pid_match, data); + return io_wq_cancel_cb(wq, io_wq_io_cb_cancel_data, (void *)cwork, false); } struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) diff --git a/fs/io-wq.h b/fs/io-wq.h index 8e138fa88b9f..071f1a997800 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -90,7 +90,6 @@ struct io_wq_work { const struct cred *creds; struct fs_struct *fs; unsigned flags; - pid_t task_pid; }; static inline struct io_wq_work *wq_next_work(struct io_wq_work *work) @@ -125,12 +124,11 @@ static inline bool io_wq_is_hashed(struct io_wq_work *work) void io_wq_cancel_all(struct io_wq *wq); enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork); -enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid); typedef bool (work_cancel_fn)(struct io_wq_work *, void *); enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, - void *data); + void *data, bool cancel_all); struct task_struct *io_wq_get_task(struct io_wq *wq); diff --git a/fs/io_uring.c b/fs/io_uring.c index 155f3d830ddb..a78201b96179 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -541,6 +541,7 @@ enum { REQ_F_NO_FILE_TABLE_BIT, REQ_F_QUEUE_TIMEOUT_BIT, REQ_F_WORK_INITIALIZED_BIT, + REQ_F_TASK_PINNED_BIT, /* not a real bit, just to check we're not overflowing the space */ __REQ_F_LAST_BIT, @@ -598,6 +599,8 @@ enum { REQ_F_QUEUE_TIMEOUT = BIT(REQ_F_QUEUE_TIMEOUT_BIT), /* io_wq_work is initialized */ REQ_F_WORK_INITIALIZED = BIT(REQ_F_WORK_INITIALIZED_BIT), + /* req->task is refcounted */ + REQ_F_TASK_PINNED = BIT(REQ_F_TASK_PINNED_BIT), }; struct async_poll { @@ -910,6 +913,21 @@ struct sock *io_uring_get_socket(struct file *file) } EXPORT_SYMBOL(io_uring_get_socket); +static void io_get_req_task(struct io_kiocb *req) +{ + if (req->flags & REQ_F_TASK_PINNED) + return; + get_task_struct(req->task); + req->flags |= REQ_F_TASK_PINNED; +} + +/* not idempotent -- it doesn't clear REQ_F_TASK_PINNED */ +static void __io_put_req_task(struct io_kiocb *req) +{ + if (req->flags & REQ_F_TASK_PINNED) + put_task_struct(req->task); +} + static void io_file_put_work(struct work_struct *work); /* @@ -1045,8 +1063,6 @@ static inline void io_req_work_grab_env(struct io_kiocb *req, } spin_unlock(¤t->fs->lock); } - if (!req->work.task_pid) - req->work.task_pid = task_pid_vnr(current); } static inline void io_req_work_drop_env(struct io_kiocb *req) @@ -1087,6 +1103,7 @@ static inline void io_prep_async_work(struct io_kiocb *req, req->work.flags |= IO_WQ_WORK_UNBOUND; } + io_req_init_async(req); io_req_work_grab_env(req, def); *link = io_prep_linked_timeout(req); @@ -1398,9 +1415,7 @@ static void __io_req_aux_free(struct io_kiocb *req) kfree(req->io); if (req->file) io_put_file(req, req->file, (req->flags & REQ_F_FIXED_FILE)); - if (req->task) - put_task_struct(req->task); - + __io_put_req_task(req); io_req_work_drop_env(req); } @@ -1727,6 +1742,18 @@ static int io_put_kbuf(struct io_kiocb *req) return cflags; } +static void io_iopoll_queue(struct list_head *again) +{ + struct io_kiocb *req; + + do { + req = list_first_entry(again, struct io_kiocb, list); + list_del(&req->list); + refcount_inc(&req->refs); + io_queue_async_work(req); + } while (!list_empty(again)); +} + /* * Find and free completed poll iocbs */ @@ -1735,12 +1762,21 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, { struct req_batch rb; struct io_kiocb *req; + LIST_HEAD(again); + + /* order with ->result store in io_complete_rw_iopoll() */ + smp_rmb(); rb.to_free = rb.need_iter = 0; while (!list_empty(done)) { int cflags = 0; req = list_first_entry(done, struct io_kiocb, list); + if (READ_ONCE(req->result) == -EAGAIN) { + req->iopoll_completed = 0; + list_move_tail(&req->list, &again); + continue; + } list_del(&req->list); if (req->flags & REQ_F_BUFFER_SELECTED) @@ -1758,18 +1794,9 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, if (ctx->flags & IORING_SETUP_SQPOLL) io_cqring_ev_posted(ctx); io_free_req_many(ctx, &rb); -} -static void io_iopoll_queue(struct list_head *again) -{ - struct io_kiocb *req; - - do { - req = list_first_entry(again, struct io_kiocb, list); - list_del(&req->list); - refcount_inc(&req->refs); - io_queue_async_work(req); - } while (!list_empty(again)); + if (!list_empty(&again)) + io_iopoll_queue(&again); } static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events, @@ -1777,7 +1804,6 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events, { struct io_kiocb *req, *tmp; LIST_HEAD(done); - LIST_HEAD(again); bool spin; int ret; @@ -1803,13 +1829,6 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events, if (!list_empty(&done)) break; - if (req->result == -EAGAIN) { - list_move_tail(&req->list, &again); - continue; - } - if (!list_empty(&again)) - break; - ret = kiocb->ki_filp->f_op->iopoll(kiocb, spin); if (ret < 0) break; @@ -1822,9 +1841,6 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events, if (!list_empty(&done)) io_iopoll_complete(ctx, nr_events, &done); - if (!list_empty(&again)) - io_iopoll_queue(&again); - return ret; } @@ -1973,11 +1989,15 @@ static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2) if (kiocb->ki_flags & IOCB_WRITE) kiocb_end_write(req); - if (res != req->result) + if (res != -EAGAIN && res != req->result) req_set_fail_links(req); - req->result = res; - if (res != -EAGAIN) + + WRITE_ONCE(req->result, res); + /* order with io_poll_complete() checking ->result */ + if (res != -EAGAIN) { + smp_wmb(); WRITE_ONCE(req->iopoll_completed, 1); + } } /* @@ -2650,8 +2670,8 @@ copy_iov: } } out_free: - kfree(iovec); - req->flags &= ~REQ_F_NEED_CLEANUP; + if (!(req->flags & REQ_F_NEED_CLEANUP)) + kfree(iovec); return ret; } @@ -2773,8 +2793,8 @@ copy_iov: } } out_free: - req->flags &= ~REQ_F_NEED_CLEANUP; - kfree(iovec); + if (!(req->flags & REQ_F_NEED_CLEANUP)) + kfree(iovec); return ret; } @@ -4236,6 +4256,28 @@ static void io_async_queue_proc(struct file *file, struct wait_queue_head *head, __io_queue_proc(&pt->req->apoll->poll, pt, head); } +static void io_sq_thread_drop_mm(struct io_ring_ctx *ctx) +{ + struct mm_struct *mm = current->mm; + + if (mm) { + kthread_unuse_mm(mm); + mmput(mm); + } +} + +static int io_sq_thread_acquire_mm(struct io_ring_ctx *ctx, + struct io_kiocb *req) +{ + if (io_op_defs[req->opcode].needs_mm && !current->mm) { + if (unlikely(!mmget_not_zero(ctx->sqo_mm))) + return -EFAULT; + kthread_use_mm(ctx->sqo_mm); + } + + return 0; +} + static void io_async_task_func(struct callback_head *cb) { struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work); @@ -4270,11 +4312,16 @@ static void io_async_task_func(struct callback_head *cb) if (!canceled) { __set_current_state(TASK_RUNNING); + if (io_sq_thread_acquire_mm(ctx, req)) { + io_cqring_add_event(req, -EFAULT); + goto end_req; + } mutex_lock(&ctx->uring_lock); __io_queue_sqe(req, NULL); mutex_unlock(&ctx->uring_lock); } else { io_cqring_ev_posted(ctx); +end_req: req_set_fail_links(req); io_double_put_req(req); } @@ -4366,8 +4413,7 @@ static bool io_arm_poll_handler(struct io_kiocb *req) memcpy(&apoll->work, &req->work, sizeof(req->work)); had_io = req->io != NULL; - get_task_struct(current); - req->task = current; + io_get_req_task(req); req->apoll = apoll; INIT_HLIST_NODE(&req->hash_node); @@ -4555,8 +4601,7 @@ static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe events = READ_ONCE(sqe->poll_events); poll->events = demangle_poll(events) | EPOLLERR | EPOLLHUP; - get_task_struct(current); - req->task = current; + io_get_req_task(req); return 0; } @@ -4772,7 +4817,7 @@ static int io_async_cancel_one(struct io_ring_ctx *ctx, void *sqe_addr) enum io_wq_cancel cancel_ret; int ret = 0; - cancel_ret = io_wq_cancel_cb(ctx->io_wq, io_cancel_cb, sqe_addr); + cancel_ret = io_wq_cancel_cb(ctx->io_wq, io_cancel_cb, sqe_addr, false); switch (cancel_ret) { case IO_WQ_CANCEL_OK: ret = 0; @@ -5817,17 +5862,14 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, req->flags = 0; /* one is dropped after submission, the other at completion */ refcount_set(&req->refs, 2); - req->task = NULL; + req->task = current; req->result = 0; if (unlikely(req->opcode >= IORING_OP_LAST)) return -EINVAL; - if (io_op_defs[req->opcode].needs_mm && !current->mm) { - if (unlikely(!mmget_not_zero(ctx->sqo_mm))) - return -EFAULT; - kthread_use_mm(ctx->sqo_mm); - } + if (unlikely(io_sq_thread_acquire_mm(ctx, req))) + return -EFAULT; sqe_flags = READ_ONCE(sqe->flags); /* enforce forwards compatibility on users */ @@ -5936,16 +5978,6 @@ fail_req: return submitted; } -static inline void io_sq_thread_drop_mm(struct io_ring_ctx *ctx) -{ - struct mm_struct *mm = current->mm; - - if (mm) { - kthread_unuse_mm(mm); - mmput(mm); - } -} - static int io_sq_thread(void *data) { struct io_ring_ctx *ctx = data; @@ -7331,7 +7363,17 @@ static void io_ring_exit_work(struct work_struct *work) if (ctx->rings) io_cqring_overflow_flush(ctx, true); - wait_for_completion(&ctx->ref_comp); + /* + * If we're doing polled IO and end up having requests being + * submitted async (out-of-line), then completions can come in while + * we're waiting for refs to drop. We need to reap these manually, + * as nobody else will be looking for them. + */ + while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)) { + io_iopoll_reap_events(ctx); + if (ctx->rings) + io_cqring_overflow_flush(ctx, true); + } io_ring_ctx_free(ctx); } @@ -7365,9 +7407,22 @@ static int io_uring_release(struct inode *inode, struct file *file) return 0; } +static bool io_wq_files_match(struct io_wq_work *work, void *data) +{ + struct files_struct *files = data; + + return work->files == files; +} + static void io_uring_cancel_files(struct io_ring_ctx *ctx, struct files_struct *files) { + if (list_empty_careful(&ctx->inflight_list)) + return; + + /* cancel all at once, should be faster than doing it one by one*/ + io_wq_cancel_cb(ctx->io_wq, io_wq_files_match, files, true); + while (!list_empty_careful(&ctx->inflight_list)) { struct io_kiocb *cancel_req = NULL, *req; DEFINE_WAIT(wait); @@ -7423,6 +7478,14 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx, } } +static bool io_cancel_task_cb(struct io_wq_work *work, void *data) +{ + struct io_kiocb *req = container_of(work, struct io_kiocb, work); + struct task_struct *task = data; + + return req->task == task; +} + static int io_uring_flush(struct file *file, void *data) { struct io_ring_ctx *ctx = file->private_data; @@ -7433,7 +7496,7 @@ static int io_uring_flush(struct file *file, void *data) * If the task is going away, cancel work it may have pending */ if (fatal_signal_pending(current) || (current->flags & PF_EXITING)) - io_wq_cancel_pid(ctx->io_wq, task_pid_vnr(current)); + io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, current, true); return 0; } diff --git a/fs/proc/bootconfig.c b/fs/proc/bootconfig.c index 9955d75c0585..ad31ec4ad627 100644 --- a/fs/proc/bootconfig.c +++ b/fs/proc/bootconfig.c @@ -26,8 +26,9 @@ static int boot_config_proc_show(struct seq_file *m, void *v) static int __init copy_xbc_key_value_list(char *dst, size_t size) { struct xbc_node *leaf, *vnode; - const char *val; char *key, *end = dst + size; + const char *val; + char q; int ret = 0; key = kzalloc(XBC_KEYLEN_MAX, GFP_KERNEL); @@ -41,16 +42,20 @@ static int __init copy_xbc_key_value_list(char *dst, size_t size) break; dst += ret; vnode = xbc_node_get_child(leaf); - if (vnode && xbc_node_is_array(vnode)) { + if (vnode) { xbc_array_for_each_value(vnode, val) { - ret = snprintf(dst, rest(dst, end), "\"%s\"%s", - val, vnode->next ? ", " : "\n"); + if (strchr(val, '"')) + q = '\''; + else + q = '"'; + ret = snprintf(dst, rest(dst, end), "%c%s%c%s", + q, val, q, vnode->next ? ", " : "\n"); if (ret < 0) goto out; dst += ret; } } else { - ret = snprintf(dst, rest(dst, end), "\"%s\"\n", val); + ret = snprintf(dst, rest(dst, end), "\"\"\n"); if (ret < 0) break; dst += ret; diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index 8ba492d44e68..e502414b3556 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -512,7 +512,8 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos) * Using bounce buffer to bypass the * hardened user copy kernel text checks. */ - if (probe_kernel_read(buf, (void *) start, tsz)) { + if (copy_from_kernel_nofault(buf, (void *)start, + tsz)) { if (clear_user(buffer, tsz)) { ret = -EFAULT; goto out; |