diff options
Diffstat (limited to 'fs/btrfs/inode.c')
-rw-r--r-- | fs/btrfs/inode.c | 99 |
1 files changed, 60 insertions, 39 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 5136c73b3dce..f751ab47e9a5 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3080,56 +3080,46 @@ static int btrfs_readpage_end_io_hook(struct btrfs_io_bio *io_bio, start, (size_t)(end - start + 1)); } -struct delayed_iput { - struct list_head list; - struct inode *inode; -}; - -/* JDM: If this is fs-wide, why can't we add a pointer to - * btrfs_inode instead and avoid the allocation? */ void btrfs_add_delayed_iput(struct inode *inode) { struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; - struct delayed_iput *delayed; + struct btrfs_inode *binode = BTRFS_I(inode); if (atomic_add_unless(&inode->i_count, -1, 1)) return; - delayed = kmalloc(sizeof(*delayed), GFP_NOFS | __GFP_NOFAIL); - delayed->inode = inode; - spin_lock(&fs_info->delayed_iput_lock); - list_add_tail(&delayed->list, &fs_info->delayed_iputs); + if (binode->delayed_iput_count == 0) { + ASSERT(list_empty(&binode->delayed_iput)); + list_add_tail(&binode->delayed_iput, &fs_info->delayed_iputs); + } else { + binode->delayed_iput_count++; + } spin_unlock(&fs_info->delayed_iput_lock); } void btrfs_run_delayed_iputs(struct btrfs_root *root) { - LIST_HEAD(list); struct btrfs_fs_info *fs_info = root->fs_info; - struct delayed_iput *delayed; - int empty; - - spin_lock(&fs_info->delayed_iput_lock); - empty = list_empty(&fs_info->delayed_iputs); - spin_unlock(&fs_info->delayed_iput_lock); - if (empty) - return; - - down_read(&fs_info->delayed_iput_sem); spin_lock(&fs_info->delayed_iput_lock); - list_splice_init(&fs_info->delayed_iputs, &list); - spin_unlock(&fs_info->delayed_iput_lock); - - while (!list_empty(&list)) { - delayed = list_entry(list.next, struct delayed_iput, list); - list_del(&delayed->list); - iput(delayed->inode); - kfree(delayed); + while (!list_empty(&fs_info->delayed_iputs)) { + struct btrfs_inode *inode; + + inode = list_first_entry(&fs_info->delayed_iputs, + struct btrfs_inode, delayed_iput); + if (inode->delayed_iput_count) { + inode->delayed_iput_count--; + list_move_tail(&inode->delayed_iput, + &fs_info->delayed_iputs); + } else { + list_del_init(&inode->delayed_iput); + } + spin_unlock(&fs_info->delayed_iput_lock); + iput(&inode->vfs_inode); + spin_lock(&fs_info->delayed_iput_lock); } - - up_read(&root->fs_info->delayed_iput_sem); + spin_unlock(&fs_info->delayed_iput_lock); } /* @@ -5676,6 +5666,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) char *name_ptr; int name_len; int is_curr = 0; /* ctx->pos points to the current index? */ + bool emitted; /* FIXME, use a real flag for deciding about the key type */ if (root->fs_info->tree_root == root) @@ -5704,6 +5695,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx) if (ret < 0) goto err; + emitted = false; while (1) { leaf = path->nodes[0]; slot = path->slots[0]; @@ -5783,6 +5775,7 @@ skip: if (over) goto nopos; + emitted = true; di_len = btrfs_dir_name_len(leaf, di) + btrfs_dir_data_len(leaf, di) + sizeof(*di); di_cur += di_len; @@ -5795,11 +5788,20 @@ next: if (key_type == BTRFS_DIR_INDEX_KEY) { if (is_curr) ctx->pos++; - ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list); + ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list, &emitted); if (ret) goto nopos; } + /* + * If we haven't emitted any dir entry, we must not touch ctx->pos as + * it was was set to the termination value in previous call. We assume + * that "." and ".." were emitted if we reach this point and set the + * termination value as well for an empty directory. + */ + if (ctx->pos > 2 && !emitted) + goto nopos; + /* Reached end of directory/root. Bump pos past the last item. */ ctx->pos++; @@ -6419,7 +6421,7 @@ out_unlock_inode: static int btrfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) { - struct btrfs_trans_handle *trans; + struct btrfs_trans_handle *trans = NULL; struct btrfs_root *root = BTRFS_I(dir)->root; struct inode *inode = d_inode(old_dentry); u64 index; @@ -6445,6 +6447,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, trans = btrfs_start_transaction(root, 5); if (IS_ERR(trans)) { err = PTR_ERR(trans); + trans = NULL; goto fail; } @@ -6478,9 +6481,10 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, btrfs_log_new_name(trans, inode, NULL, parent); } - btrfs_end_transaction(trans, root); btrfs_balance_delayed_items(root); fail: + if (trans) + btrfs_end_transaction(trans, root); if (drop_inode) { inode_dec_link_count(inode); iput(inode); @@ -8414,15 +8418,28 @@ int btrfs_readpage(struct file *file, struct page *page) static int btrfs_writepage(struct page *page, struct writeback_control *wbc) { struct extent_io_tree *tree; - + struct inode *inode = page->mapping->host; + int ret; if (current->flags & PF_MEMALLOC) { redirty_page_for_writepage(wbc, page); unlock_page(page); return 0; } + + /* + * If we are under memory pressure we will call this directly from the + * VM, we need to make sure we have the inode referenced for the ordered + * extent. If not just return like we didn't do anything. + */ + if (!igrab(inode)) { + redirty_page_for_writepage(wbc, page); + return AOP_WRITEPAGE_ACTIVATE; + } tree = &BTRFS_I(page->mapping->host)->io_tree; - return extent_write_full_page(tree, page, btrfs_get_extent, wbc); + ret = extent_write_full_page(tree, page, btrfs_get_extent, wbc); + btrfs_add_delayed_iput(inode); + return ret; } static int btrfs_writepages(struct address_space *mapping, @@ -8890,6 +8907,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->dir_index = 0; ei->last_unlink_trans = 0; ei->last_log_commit = 0; + ei->delayed_iput_count = 0; spin_lock_init(&ei->lock); ei->outstanding_extents = 0; @@ -8914,6 +8932,7 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) mutex_init(&ei->delalloc_mutex); btrfs_ordered_inode_tree_init(&ei->ordered_tree); INIT_LIST_HEAD(&ei->delalloc_inodes); + INIT_LIST_HEAD(&ei->delayed_iput); RB_CLEAR_NODE(&ei->rb_node); return inode; @@ -9501,9 +9520,11 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, /* * 2 items for inode item and ref * 2 items for dir items + * 1 item for updating parent inode item + * 1 item for the inline extent item * 1 item for xattr if selinux is on */ - trans = btrfs_start_transaction(root, 5); + trans = btrfs_start_transaction(root, 7); if (IS_ERR(trans)) return PTR_ERR(trans); |