diff options
Diffstat (limited to 'fs/btrfs/ioctl.c')
| -rw-r--r-- | fs/btrfs/ioctl.c | 641 |
1 files changed, 554 insertions, 87 deletions
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index a6d8efa46bf..82c18ba12e3 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -58,6 +58,33 @@ #include "dev-replace.h" #include "props.h" #include "sysfs.h" +#include "qgroup.h" + +#ifdef CONFIG_64BIT +/* If we have a 32-bit userspace and 64-bit kernel, then the UAPI + * structures are incorrect, as the timespec structure from userspace + * is 4 bytes too small. We define these alternatives here to teach + * the kernel about the 32-bit struct packing. + */ +struct btrfs_ioctl_timespec_32 { + __u64 sec; + __u32 nsec; +} __attribute__ ((__packed__)); + +struct btrfs_ioctl_received_subvol_args_32 { + char uuid[BTRFS_UUID_SIZE]; /* in */ + __u64 stransid; /* in */ + __u64 rtransid; /* out */ + struct btrfs_ioctl_timespec_32 stime; /* in */ + struct btrfs_ioctl_timespec_32 rtime; /* out */ + __u64 flags; /* in */ + __u64 reserved[16]; /* in */ +} __attribute__ ((__packed__)); + +#define BTRFS_IOC_SET_RECEIVED_SUBVOL_32 _IOWR(BTRFS_IOCTL_MAGIC, 37, \ + struct btrfs_ioctl_received_subvol_args_32) +#endif + static int btrfs_clone(struct inode *src, struct inode *inode, u64 off, u64 olen, u64 olen_aligned, u64 destoff); @@ -585,6 +612,23 @@ fail: return ret; } +static void btrfs_wait_nocow_write(struct btrfs_root *root) +{ + s64 writers; + DEFINE_WAIT(wait); + + do { + prepare_to_wait(&root->subv_writers->wait, &wait, + TASK_UNINTERRUPTIBLE); + + writers = percpu_counter_sum(&root->subv_writers->counter); + if (writers) + schedule(); + + finish_wait(&root->subv_writers->wait, &wait); + } while (writers); +} + static int create_snapshot(struct btrfs_root *root, struct inode *dir, struct dentry *dentry, char *name, int namelen, u64 *async_transid, bool readonly, @@ -595,18 +639,24 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, struct btrfs_trans_handle *trans; int ret; - if (!root->ref_cows) + if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state)) return -EINVAL; + atomic_inc(&root->will_be_snapshoted); + smp_mb__after_atomic(); + btrfs_wait_nocow_write(root); + ret = btrfs_start_delalloc_inodes(root, 0); if (ret) - return ret; + goto out; btrfs_wait_ordered_extents(root, -1); pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS); - if (!pending_snapshot) - return -ENOMEM; + if (!pending_snapshot) { + ret = -ENOMEM; + goto out; + } btrfs_init_block_rsv(&pending_snapshot->block_rsv, BTRFS_BLOCK_RSV_TEMP); @@ -623,7 +673,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, &pending_snapshot->qgroup_reserved, false); if (ret) - goto out; + goto free; pending_snapshot->dentry = dentry; pending_snapshot->root = root; @@ -662,6 +712,35 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, if (ret) goto fail; + /* + * If orphan cleanup did remove any orphans, it means the tree was + * modified and therefore the commit root is not the same as the + * current root anymore. This is a problem, because send uses the + * commit root and therefore can see inode items that don't exist + * in the current root anymore, and for example make calls to + * btrfs_iget, which will do tree lookups based on the current root + * and not on the commit root. Those lookups will fail, returning a + * -ESTALE error, and making send fail with that error. So make sure + * a send does not see any orphans we have just removed, and that it + * will see the same inodes regardless of whether a transaction + * commit happened before it started (meaning that the commit root + * will be the same as the current root) or not. + */ + if (readonly && pending_snapshot->snap->node != + pending_snapshot->snap->commit_root) { + trans = btrfs_join_transaction(pending_snapshot->snap); + if (IS_ERR(trans) && PTR_ERR(trans) != -ENOENT) { + ret = PTR_ERR(trans); + goto fail; + } + if (!IS_ERR(trans)) { + ret = btrfs_commit_transaction(trans, + pending_snapshot->snap); + if (ret) + goto fail; + } + } + inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry); if (IS_ERR(inode)) { ret = PTR_ERR(inode); @@ -674,8 +753,10 @@ fail: btrfs_subvolume_release_metadata(BTRFS_I(dir)->root, &pending_snapshot->block_rsv, pending_snapshot->qgroup_reserved); -out: +free: kfree(pending_snapshot); +out: + atomic_dec(&root->will_be_snapshoted); return ret; } @@ -884,12 +965,14 @@ static int find_new_extents(struct btrfs_root *root, min_key.type = BTRFS_EXTENT_DATA_KEY; min_key.offset = *off; - path->keep_locks = 1; - while (1) { + path->keep_locks = 1; ret = btrfs_search_forward(root, &min_key, path, newer_than); if (ret != 0) goto none; + path->keep_locks = 0; + btrfs_unlock_up_safe(path, 1); +process_slot: if (min_key.objectid != ino) goto none; if (min_key.type != BTRFS_EXTENT_DATA_KEY) @@ -908,6 +991,12 @@ static int find_new_extents(struct btrfs_root *root, return 0; } + path->slots[0]++; + if (path->slots[0] < btrfs_header_nritems(leaf)) { + btrfs_item_key_to_cpu(leaf, &min_key, path->slots[0]); + goto process_slot; + } + if (min_key.offset == (u64)-1) goto none; @@ -935,10 +1024,13 @@ static struct extent_map *defrag_lookup_extent(struct inode *inode, u64 start) read_unlock(&em_tree->lock); if (!em) { + struct extent_state *cached = NULL; + u64 end = start + len - 1; + /* get the big lock and read metadata off disk */ - lock_extent(io_tree, start, start + len - 1); + lock_extent_bits(io_tree, start, end, 0, &cached); em = btrfs_get_extent(inode, NULL, 0, start, len, 0); - unlock_extent(io_tree, start, start + len - 1); + unlock_extent_cached(io_tree, start, end, &cached, GFP_NOFS); if (IS_ERR(em)) return NULL; @@ -957,7 +1049,8 @@ static bool defrag_check_next_extent(struct inode *inode, struct extent_map *em) return false; next = defrag_lookup_extent(inode, em->start + em->len); - if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE) + if (!next || next->block_start >= EXTENT_MAP_LAST_BYTE || + (em->block_start + em->block_len == next->block_start)) ret = false; free_extent_map(next); @@ -1076,10 +1169,12 @@ again: page_start = page_offset(page); page_end = page_start + PAGE_CACHE_SIZE - 1; while (1) { - lock_extent(tree, page_start, page_end); + lock_extent_bits(tree, page_start, page_end, + 0, &cached_state); ordered = btrfs_lookup_ordered_extent(inode, page_start); - unlock_extent(tree, page_start, page_end); + unlock_extent_cached(tree, page_start, page_end, + &cached_state, GFP_NOFS); if (!ordered) break; @@ -1356,8 +1451,12 @@ int btrfs_defrag_file(struct inode *inode, struct file *file, } } - if ((range->flags & BTRFS_DEFRAG_RANGE_START_IO)) + if ((range->flags & BTRFS_DEFRAG_RANGE_START_IO)) { filemap_flush(inode->i_mapping); + if (test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, + &BTRFS_I(inode)->runtime_flags)) + filemap_flush(inode->i_mapping); + } if ((range->flags & BTRFS_DEFRAG_RANGE_COMPRESS)) { /* the filemap_flush will queue IO into the worker threads, but @@ -1403,6 +1502,7 @@ static noinline int btrfs_ioctl_resize(struct file *file, struct btrfs_trans_handle *trans; struct btrfs_device *device = NULL; char *sizestr; + char *retptr; char *devstr = NULL; int ret = 0; int mod = 0; @@ -1432,11 +1532,12 @@ static noinline int btrfs_ioctl_resize(struct file *file, sizestr = vol_args->name; devstr = strchr(sizestr, ':'); if (devstr) { - char *end; sizestr = devstr + 1; *devstr = '\0'; devstr = vol_args->name; - devid = simple_strtoull(devstr, &end, 10); + ret = kstrtoull(devstr, 10, &devid); + if (ret) + goto out_free; if (!devid) { ret = -EINVAL; goto out_free; @@ -1470,8 +1571,8 @@ static noinline int btrfs_ioctl_resize(struct file *file, mod = 1; sizestr++; } - new_size = memparse(sizestr, NULL); - if (new_size == 0) { + new_size = memparse(sizestr, &retptr); + if (*retptr != '\0' || new_size == 0) { ret = -EINVAL; goto out_free; } @@ -1492,7 +1593,7 @@ static noinline int btrfs_ioctl_resize(struct file *file, new_size = old_size - new_size; } else if (mod > 0) { if (new_size > ULLONG_MAX - old_size) { - ret = -EINVAL; + ret = -ERANGE; goto out_free; } new_size = old_size + new_size; @@ -1573,7 +1674,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, if (src_inode->i_sb != file_inode(file)->i_sb) { btrfs_info(BTRFS_I(src_inode)->root->fs_info, "Snapshot src from another FS"); - ret = -EINVAL; + ret = -EXDEV; } else if (!inode_owner_or_capable(src_inode)) { /* * Subvolume creation is not restricted, but snapshots @@ -1797,7 +1898,9 @@ static noinline int may_destroy_subvol(struct btrfs_root *root) if (di && !IS_ERR(di)) { btrfs_dir_item_key_to_cpu(path->nodes[0], di, &key); if (key.objectid == root->root_key.objectid) { - ret = -ENOTEMPTY; + ret = -EPERM; + btrfs_err(root->fs_info, "deleting default subvolume " + "%llu is not allowed", key.objectid); goto out; } btrfs_release_path(path); @@ -2147,6 +2250,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, struct btrfs_ioctl_vol_args *vol_args; struct btrfs_trans_handle *trans; struct btrfs_block_rsv block_rsv; + u64 root_flags; u64 qgroup_reserved; int namelen; int ret; @@ -2168,6 +2272,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, if (err) goto out; + err = mutex_lock_killable_nested(&dir->i_mutex, I_MUTEX_PARENT); if (err == -EINTR) goto out_drop_write; @@ -2229,6 +2334,27 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, } mutex_lock(&inode->i_mutex); + + /* + * Don't allow to delete a subvolume with send in progress. This is + * inside the i_mutex so the error handling that has to drop the bit + * again is not run concurrently. + */ + spin_lock(&dest->root_item_lock); + root_flags = btrfs_root_flags(&dest->root_item); + if (dest->send_in_progress == 0) { + btrfs_set_root_flags(&dest->root_item, + root_flags | BTRFS_ROOT_SUBVOL_DEAD); + spin_unlock(&dest->root_item_lock); + } else { + spin_unlock(&dest->root_item_lock); + btrfs_warn(root->fs_info, + "Attempt to delete subvolume %llu during send", + dest->root_key.objectid); + err = -EPERM; + goto out_dput; + } + err = d_invalidate(dentry); if (err) goto out_unlock; @@ -2274,7 +2400,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, dest->root_item.drop_level = 0; btrfs_set_root_refs(&dest->root_item, 0); - if (!xchg(&dest->orphan_item_inserted, 1)) { + if (!test_and_set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &dest->state)) { ret = btrfs_insert_orphan_item(trans, root->fs_info->tree_root, dest->root_key.objectid); @@ -2317,11 +2443,19 @@ out_release: out_up_write: up_write(&root->fs_info->subvol_sem); out_unlock: + if (err) { + spin_lock(&dest->root_item_lock); + root_flags = btrfs_root_flags(&dest->root_item); + btrfs_set_root_flags(&dest->root_item, + root_flags & ~BTRFS_ROOT_SUBVOL_DEAD); + spin_unlock(&dest->root_item_lock); + } mutex_unlock(&inode->i_mutex); if (!err) { shrink_dcache_sb(root->fs_info->sb); btrfs_invalidate_inodes(dest); d_delete(dentry); + ASSERT(dest->send_in_progress == 0); /* the last ref */ if (dest->cache_inode) { @@ -2485,9 +2619,6 @@ static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg) struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices; int ret = 0; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - fi_args = kzalloc(sizeof(*fi_args), GFP_KERNEL); if (!fi_args) return -ENOMEM; @@ -2502,6 +2633,10 @@ static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg) } mutex_unlock(&fs_devices->device_list_mutex); + fi_args->nodesize = root->fs_info->super_copy->nodesize; + fi_args->sectorsize = root->fs_info->super_copy->sectorsize; + fi_args->clone_alignment = root->fs_info->super_copy->sectorsize; + if (copy_to_user(arg, fi_args, sizeof(*fi_args))) ret = -EFAULT; @@ -2517,9 +2652,6 @@ static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg) int ret = 0; char *s_uuid = NULL; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - di_args = memdup_user(arg, sizeof(*di_args)); if (IS_ERR(di_args)) return PTR_ERR(di_args); @@ -2597,10 +2729,15 @@ static inline void lock_extent_range(struct inode *inode, u64 off, u64 len) lock_extent(&BTRFS_I(inode)->io_tree, off, off + len - 1); ordered = btrfs_lookup_first_ordered_extent(inode, off + len - 1); - if (!ordered && + if ((!ordered || + ordered->file_offset + ordered->len <= off || + ordered->file_offset >= off + len) && !test_range_bit(&BTRFS_I(inode)->io_tree, off, - off + len - 1, EXTENT_DELALLOC, 0, NULL)) + off + len - 1, EXTENT_DELALLOC, 0, NULL)) { + if (ordered) + btrfs_put_ordered_extent(ordered); break; + } unlock_extent(&BTRFS_I(inode)->io_tree, off, off + len - 1); if (ordered) btrfs_put_ordered_extent(ordered); @@ -2840,6 +2977,126 @@ out: return ret; } +/* Helper to check and see if this root currently has a ref on the given disk + * bytenr. If it does then we need to update the quota for this root. This + * doesn't do anything if quotas aren't enabled. + */ +static int check_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, + u64 disko) +{ + struct seq_list tree_mod_seq_elem = {}; + struct ulist *roots; + struct ulist_iterator uiter; + struct ulist_node *root_node = NULL; + int ret; + + if (!root->fs_info->quota_enabled) + return 1; + + btrfs_get_tree_mod_seq(root->fs_info, &tree_mod_seq_elem); + ret = btrfs_find_all_roots(trans, root->fs_info, disko, + tree_mod_seq_elem.seq, &roots); + if (ret < 0) + goto out; + ret = 0; + ULIST_ITER_INIT(&uiter); + while ((root_node = ulist_next(roots, &uiter))) { + if (root_node->val == root->objectid) { + ret = 1; + break; + } + } + ulist_free(roots); +out: + btrfs_put_tree_mod_seq(root->fs_info, &tree_mod_seq_elem); + return ret; +} + +static int clone_finish_inode_update(struct btrfs_trans_handle *trans, + struct inode *inode, + u64 endoff, + const u64 destoff, + const u64 olen) +{ + struct btrfs_root *root = BTRFS_I(inode)->root; + int ret; + + inode_inc_iversion(inode); + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + /* + * We round up to the block size at eof when determining which + * extents to clone above, but shouldn't round up the file size. + */ + if (endoff > destoff + olen) + endoff = destoff + olen; + if (endoff > inode->i_size) + btrfs_i_size_write(inode, endoff); + + ret = btrfs_update_inode(trans, root, inode); + if (ret) { + btrfs_abort_transaction(trans, root, ret); + btrfs_end_transaction(trans, root); + goto out; + } + ret = btrfs_end_transaction(trans, root); +out: + return ret; +} + +static void clone_update_extent_map(struct inode *inode, + const struct btrfs_trans_handle *trans, + const struct btrfs_path *path, + struct btrfs_file_extent_item *fi, + const u64 hole_offset, + const u64 hole_len) +{ + struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; + struct extent_map *em; + int ret; + + em = alloc_extent_map(); + if (!em) { + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + return; + } + + if (fi) { + btrfs_extent_item_to_extent_map(inode, path, fi, false, em); + em->generation = -1; + if (btrfs_file_extent_type(path->nodes[0], fi) == + BTRFS_FILE_EXTENT_INLINE) + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + } else { + em->start = hole_offset; + em->len = hole_len; + em->ram_bytes = em->len; + em->orig_start = hole_offset; + em->block_start = EXTENT_MAP_HOLE; + em->block_len = 0; + em->orig_block_len = 0; + em->compress_type = BTRFS_COMPRESS_NONE; + em->generation = trans->transid; + } + + while (1) { + write_lock(&em_tree->lock); + ret = add_extent_mapping(em_tree, em, 1); + write_unlock(&em_tree->lock); + if (ret != -EEXIST) { + free_extent_map(em); + break; + } + btrfs_drop_extent_cache(inode, em->start, + em->start + em->len - 1, 0); + } + + if (unlikely(ret)) + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); +} + /** * btrfs_clone() - clone a range from inode file to another * @@ -2852,7 +3109,8 @@ out: * @destoff: Offset within @inode to start clone */ static int btrfs_clone(struct inode *src, struct inode *inode, - u64 off, u64 olen, u64 olen_aligned, u64 destoff) + const u64 off, const u64 olen, const u64 olen_aligned, + const u64 destoff) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_path *path = NULL; @@ -2863,7 +3121,10 @@ static int btrfs_clone(struct inode *src, struct inode *inode, u32 nritems; int slot; int ret; - u64 len = olen_aligned; + int no_quota; + const u64 len = olen_aligned; + u64 last_disko = 0; + u64 last_dest_end = destoff; ret = -ENOMEM; buf = vmalloc(btrfs_level_size(root, 0)); @@ -2880,7 +3141,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode, /* clone data */ key.objectid = btrfs_ino(src); key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = 0; + key.offset = off; while (1) { /* @@ -2892,9 +3153,21 @@ static int btrfs_clone(struct inode *src, struct inode *inode, 0, 0); if (ret < 0) goto out; + /* + * First search, if no extent item that starts at offset off was + * found but the previous item is an extent item, it's possible + * it might overlap our target range, therefore process it. + */ + if (key.offset == off && ret > 0 && path->slots[0] > 0) { + btrfs_item_key_to_cpu(path->nodes[0], &key, + path->slots[0] - 1); + if (key.type == BTRFS_EXTENT_DATA_KEY) + path->slots[0]--; + } nritems = btrfs_header_nritems(path->nodes[0]); process_slot: + no_quota = 1; if (path->slots[0] >= nritems) { ret = btrfs_next_leaf(BTRFS_I(src)->root, path); if (ret < 0) @@ -2919,7 +3192,7 @@ process_slot: u64 disko = 0, diskl = 0; u64 datao = 0, datal = 0; u8 comp; - u64 endoff; + u64 drop_start; extent = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); @@ -2940,10 +3213,16 @@ process_slot: extent); } - if (key.offset + datal <= off || - key.offset >= off + len - 1) { + /* + * The first search might have left us at an extent + * item that ends before our target range's start, can + * happen if we have holes and NO_HOLES feature enabled. + */ + if (key.offset + datal <= off) { path->slots[0]++; goto process_slot; + } else if (key.offset >= off + len) { + break; } size = btrfs_item_size_nr(leaf, slot); @@ -2962,6 +3241,18 @@ process_slot: new_key.offset = destoff; /* + * Deal with a hole that doesn't have an extent item + * that represents it (NO_HOLES feature enabled). + * This hole is either in the middle of the cloning + * range or at the beginning (fully overlaps it or + * partially overlaps it). + */ + if (new_key.offset != last_dest_end) + drop_start = last_dest_end; + else + drop_start = new_key.offset; + + /* * 1 - adjusting old extent (we may have to split it) * 1 - add new extent * 1 - inode update @@ -2979,23 +3270,24 @@ process_slot: * | ------------- extent ------------- | */ - /* substract range b */ + /* subtract range b */ if (key.offset + datal > off + len) datal = off + len - key.offset; - /* substract range a */ + /* subtract range a */ if (off > key.offset) { datao += off - key.offset; datal -= off - key.offset; } ret = btrfs_drop_extents(trans, root, inode, - new_key.offset, + drop_start, new_key.offset + datal, 1); if (ret) { - btrfs_abort_transaction(trans, root, - ret); + if (ret != -EOPNOTSUPP) + btrfs_abort_transaction(trans, + root, ret); btrfs_end_transaction(trans, root); goto out; } @@ -3026,6 +3318,28 @@ process_slot: datao); btrfs_set_file_extent_num_bytes(leaf, extent, datal); + + /* + * We need to look up the roots that point at + * this bytenr and see if the new root does. If + * it does not we need to make sure we update + * quotas appropriately. + */ + if (disko && root != BTRFS_I(src)->root && + disko != last_disko) { + no_quota = check_ref(trans, root, + disko); + if (no_quota < 0) { + btrfs_abort_transaction(trans, + root, + ret); + btrfs_end_transaction(trans, + root); + ret = no_quota; + goto out; + } + } + if (disko) { inode_add_bytes(inode, datal); ret = btrfs_inc_extent_ref(trans, root, @@ -3033,7 +3347,7 @@ process_slot: root->root_key.objectid, btrfs_ino(inode), new_key.offset - datao, - 0); + no_quota); if (ret) { btrfs_abort_transaction(trans, root, @@ -3047,6 +3361,8 @@ process_slot: } else if (type == BTRFS_FILE_EXTENT_INLINE) { u64 skip = 0; u64 trim = 0; + u64 aligned_end = 0; + if (off > key.offset) { skip = off - key.offset; new_key.offset += skip; @@ -3063,13 +3379,16 @@ process_slot: size -= skip + trim; datal -= skip + trim; + aligned_end = ALIGN(new_key.offset + datal, + root->sectorsize); ret = btrfs_drop_extents(trans, root, inode, - new_key.offset, - new_key.offset + datal, + drop_start, + aligned_end, 1); if (ret) { - btrfs_abort_transaction(trans, root, - ret); + if (ret != -EOPNOTSUPP) + btrfs_abort_transaction(trans, + root, ret); btrfs_end_transaction(trans, root); goto out; } @@ -3096,40 +3415,69 @@ process_slot: btrfs_item_ptr_offset(leaf, slot), size); inode_add_bytes(inode, datal); + extent = btrfs_item_ptr(leaf, slot, + struct btrfs_file_extent_item); } + /* If we have an implicit hole (NO_HOLES feature). */ + if (drop_start < new_key.offset) + clone_update_extent_map(inode, trans, + path, NULL, drop_start, + new_key.offset - drop_start); + + clone_update_extent_map(inode, trans, path, + extent, 0, 0); + btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); - inode_inc_iversion(inode); - inode->i_mtime = inode->i_ctime = CURRENT_TIME; - - /* - * we round up to the block size at eof when - * determining which extents to clone above, - * but shouldn't round up the file size - */ - endoff = new_key.offset + datal; - if (endoff > destoff+olen) - endoff = destoff+olen; - if (endoff > inode->i_size) - btrfs_i_size_write(inode, endoff); - - ret = btrfs_update_inode(trans, root, inode); - if (ret) { - btrfs_abort_transaction(trans, root, ret); - btrfs_end_transaction(trans, root); + last_dest_end = new_key.offset + datal; + ret = clone_finish_inode_update(trans, inode, + last_dest_end, + destoff, olen); + if (ret) goto out; - } - ret = btrfs_end_transaction(trans, root); + if (new_key.offset + datal >= destoff + len) + break; } btrfs_release_path(path); key.offset++; } ret = 0; + if (last_dest_end < destoff + len) { + /* + * We have an implicit hole (NO_HOLES feature is enabled) that + * fully or partially overlaps our cloning range at its end. + */ + btrfs_release_path(path); + + /* + * 1 - remove extent(s) + * 1 - inode update + */ + trans = btrfs_start_transaction(root, 2); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out; + } + ret = btrfs_drop_extents(trans, root, inode, + last_dest_end, destoff + len, 1); + if (ret) { + if (ret != -EOPNOTSUPP) + btrfs_abort_transaction(trans, root, ret); + btrfs_end_transaction(trans, root); + goto out; + } + ret = clone_finish_inode_update(trans, inode, destoff + len, + destoff, olen); + if (ret) + goto out; + clone_update_extent_map(inode, trans, path, NULL, last_dest_end, + destoff + len - last_dest_end); + } + out: - btrfs_release_path(path); btrfs_free_path(path); vfree(buf); return ret; @@ -3153,8 +3501,9 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, * decompress into destination's address_space (the file offset * may change, so source mapping won't do), then recompress (or * otherwise reinsert) a subrange. - * - allow ranges within the same file to be cloned (provided - * they don't overlap)? + * + * - split destination inode's inline extents. The inline extents can + * be either compressed or non-compressed. */ /* the destination must be opened for writing */ @@ -3240,15 +3589,41 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, goto out_unlock; } - /* truncate page cache pages from target inode range */ - truncate_inode_pages_range(&inode->i_data, destoff, - PAGE_CACHE_ALIGN(destoff + len) - 1); + /* + * Lock the target range too. Right after we replace the file extent + * items in the fs tree (which now point to the cloned data), we might + * have a worker replace them with extent items relative to a write + * operation that was issued before this clone operation (i.e. confront + * with inode.c:btrfs_finish_ordered_io). + */ + if (same_inode) { + u64 lock_start = min_t(u64, off, destoff); + u64 lock_len = max_t(u64, off, destoff) + len - lock_start; - lock_extent_range(src, off, len); + lock_extent_range(src, lock_start, lock_len); + } else { + lock_extent_range(src, off, len); + lock_extent_range(inode, destoff, len); + } ret = btrfs_clone(src, inode, off, olen, len, destoff); - unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); + if (same_inode) { + u64 lock_start = min_t(u64, off, destoff); + u64 lock_end = max_t(u64, off, destoff) + len - 1; + + unlock_extent(&BTRFS_I(src)->io_tree, lock_start, lock_end); + } else { + unlock_extent(&BTRFS_I(src)->io_tree, off, off + len - 1); + unlock_extent(&BTRFS_I(inode)->io_tree, destoff, + destoff + len - 1); + } + /* + * Truncate page cache pages so that future reads will see the cloned + * data immediately and not the previous data. + */ + truncate_inode_pages_range(&inode->i_data, destoff, + PAGE_CACHE_ALIGN(destoff + len) - 1); out_unlock: if (!same_inode) { if (inode < src) { @@ -3465,6 +3840,11 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) up_read(&info->groups_sem); } + /* + * Global block reserve, exported as a space_info + */ + slot_count++; + /* space_slots == 0 means they are asking for a count */ if (space_args.space_slots == 0) { space_args.total_spaces = slot_count; @@ -3523,6 +3903,21 @@ static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) up_read(&info->groups_sem); } + /* + * Add global block reserve + */ + if (slot_count) { + struct btrfs_block_rsv *block_rsv = &root->fs_info->global_block_rsv; + + spin_lock(&block_rsv->lock); + space.total_bytes = block_rsv->size; + space.used_bytes = block_rsv->size - block_rsv->reserved; + spin_unlock(&block_rsv->lock); + space.flags = BTRFS_SPACE_INFO_GLOBAL_RSV; + memcpy(dest, &space, sizeof(space)); + space_args.total_spaces++; + } + user_dest = (struct btrfs_ioctl_space_info __user *) (arg + sizeof(struct btrfs_ioctl_space_args)); @@ -4353,10 +4748,9 @@ static long btrfs_ioctl_quota_rescan_wait(struct file *file, void __user *arg) return btrfs_qgroup_wait_for_completion(root->fs_info); } -static long btrfs_ioctl_set_received_subvol(struct file *file, - void __user *arg) +static long _btrfs_ioctl_set_received_subvol(struct file *file, + struct btrfs_ioctl_received_subvol_args *sa) { - struct btrfs_ioctl_received_subvol_args *sa = NULL; struct inode *inode = file_inode(file); struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_root_item *root_item = &root->root_item; @@ -4384,13 +4778,6 @@ static long btrfs_ioctl_set_received_subvol(struct file *file, goto out; } - sa = memdup_user(arg, sizeof(*sa)); - if (IS_ERR(sa)) { - ret = PTR_ERR(sa); - sa = NULL; - goto out; - } - /* * 1 - root item * 2 - uuid items (received uuid + subvol uuid) @@ -4444,14 +4831,90 @@ static long btrfs_ioctl_set_received_subvol(struct file *file, goto out; } +out: + up_write(&root->fs_info->subvol_sem); + mnt_drop_write_file(file); + return ret; +} + +#ifdef CONFIG_64BIT +static long btrfs_ioctl_set_received_subvol_32(struct file *file, + void __user *arg) +{ + struct btrfs_ioctl_received_subvol_args_32 *args32 = NULL; + struct btrfs_ioctl_received_subvol_args *args64 = NULL; + int ret = 0; + + args32 = memdup_user(arg, sizeof(*args32)); + if (IS_ERR(args32)) { + ret = PTR_ERR(args32); + args32 = NULL; + goto out; + } + + args64 = kmalloc(sizeof(*args64), GFP_NOFS); + if (!args64) { + ret = -ENOMEM; + goto out; + } + + memcpy(args64->uuid, args32->uuid, BTRFS_UUID_SIZE); + args64->stransid = args32->stransid; + args64->rtransid = args32->rtransid; + args64->stime.sec = args32->stime.sec; + args64->stime.nsec = args32->stime.nsec; + args64->rtime.sec = args32->rtime.sec; + args64->rtime.nsec = args32->rtime.nsec; + args64->flags = args32->flags; + + ret = _btrfs_ioctl_set_received_subvol(file, args64); + if (ret) + goto out; + + memcpy(args32->uuid, args64->uuid, BTRFS_UUID_SIZE); + args32->stransid = args64->stransid; + args32->rtransid = args64->rtransid; + args32->stime.sec = args64->stime.sec; + args32->stime.nsec = args64->stime.nsec; + args32->rtime.sec = args64->rtime.sec; + args32->rtime.nsec = args64->rtime.nsec; + args32->flags = args64->flags; + + ret = copy_to_user(arg, args32, sizeof(*args32)); + if (ret) + ret = -EFAULT; + +out: + kfree(args32); + kfree(args64); + return ret; +} +#endif + +static long btrfs_ioctl_set_received_subvol(struct file *file, + void __user *arg) +{ + struct btrfs_ioctl_received_subvol_args *sa = NULL; + int ret = 0; + + sa = memdup_user(arg, sizeof(*sa)); + if (IS_ERR(sa)) { + ret = PTR_ERR(sa); + sa = NULL; + goto out; + } + + ret = _btrfs_ioctl_set_received_subvol(file, sa); + + if (ret) + goto out; + ret = copy_to_user(arg, sa, sizeof(*sa)); if (ret) ret = -EFAULT; out: kfree(sa); - up_write(&root->fs_info->subvol_sem); - mnt_drop_write_file(file); return ret; } @@ -4746,7 +5209,7 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_SYNC: { int ret; - ret = btrfs_start_delalloc_roots(root->fs_info, 0); + ret = btrfs_start_delalloc_roots(root->fs_info, 0, -1); if (ret) return ret; ret = btrfs_sync_fs(file->f_dentry->d_sb, 1); @@ -4770,6 +5233,10 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_balance_progress(root, argp); case BTRFS_IOC_SET_RECEIVED_SUBVOL: return btrfs_ioctl_set_received_subvol(file, argp); +#ifdef CONFIG_64BIT + case BTRFS_IOC_SET_RECEIVED_SUBVOL_32: + return btrfs_ioctl_set_received_subvol_32(file, argp); +#endif case BTRFS_IOC_SEND: return btrfs_ioctl_send(file, argp); case BTRFS_IOC_GET_DEV_STATS: |