From patchwork Mon Apr 19 10:34:37 2010 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yan, Zheng" X-Patchwork-Id: 93465 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by demeter.kernel.org (8.14.3/8.14.3) with ESMTP id o3JAZiGO030354 for ; Mon, 19 Apr 2010 10:35:45 GMT Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753001Ab0DSKfm (ORCPT ); Mon, 19 Apr 2010 06:35:42 -0400 Received: from rcsinet10.oracle.com ([148.87.113.121]:56936 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752886Ab0DSKfl (ORCPT ); Mon, 19 Apr 2010 06:35:41 -0400 Received: from acsinet15.oracle.com (acsinet15.oracle.com [141.146.126.227]) by rcsinet10.oracle.com (Switch-3.4.2/Switch-3.4.1) with ESMTP id o3JAZbGM003298 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Mon, 19 Apr 2010 10:35:39 GMT Received: from acsmt353.oracle.com (acsmt353.oracle.com [141.146.40.153]) by acsinet15.oracle.com (Switch-3.4.2/Switch-3.4.1) with ESMTP id o3JAZWRL012544 for ; Mon, 19 Apr 2010 10:35:32 GMT Received: from abhmt019.oracle.com by acsmt355.oracle.com with ESMTP id 170332981271673281; Mon, 19 Apr 2010 03:34:41 -0700 Received: from [141.144.144.227] (/141.144.144.227) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Mon, 19 Apr 2010 03:34:40 -0700 Message-ID: <4BCC31BD.9040905@oracle.com> Date: Mon, 19 Apr 2010 18:34:37 +0800 From: "Yan, Zheng" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.0.4-1.fc12 Thunderbird/3.0.4 MIME-Version: 1.0 To: linux-btrfs@vger.kernel.org, Chris Mason Subject: [PATCH 11/12] Btrfs: Pre-allocate space for data relocation X-Auth-Type: Internal IP X-Source-IP: acsinet15.oracle.com [141.146.126.227] X-CT-RefId: str=0001.0A090205.4BCC31FC.00B7:SCFMA922111,ss=1,fgs=0 Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org X-Greylist: IP, sender and recipient auto-whitelisted, not delayed by milter-greylist-4.2.3 (demeter.kernel.org [140.211.167.41]); Mon, 19 Apr 2010 10:35:45 +0000 (UTC) diff -urp 11/fs/btrfs/ctree.h 12/fs/btrfs/ctree.h --- 11/fs/btrfs/ctree.h 2010-04-18 10:50:26.565702000 +0800 +++ 12/fs/btrfs/ctree.h 2010-04-18 10:55:07.771696917 +0800 @@ -2406,6 +2406,9 @@ int btrfs_cont_expand(struct inode *inod int btrfs_invalidate_inodes(struct btrfs_root *root); void btrfs_add_delayed_iput(struct inode *inode); void btrfs_run_delayed_iputs(struct btrfs_root *root); +int btrfs_prealloc_file_range(struct inode *inode, int mode, + u64 start, u64 num_bytes, u64 min_size, + loff_t actual_len, u64 *alloc_hint); extern const struct dentry_operations btrfs_dentry_operations; /* ioctl.c */ diff -urp 11/fs/btrfs/inode.c 12/fs/btrfs/inode.c --- 11/fs/btrfs/inode.c 2010-04-18 11:03:00.450703636 +0800 +++ 12/fs/btrfs/inode.c 2010-04-18 11:03:39.602698880 +0800 @@ -1174,6 +1174,13 @@ out_check: num_bytes, num_bytes, type); BUG_ON(ret); + if (root->root_key.objectid == + BTRFS_DATA_RELOC_TREE_OBJECTID) { + ret = btrfs_reloc_clone_csums(inode, cur_offset, + num_bytes); + BUG_ON(ret); + } + extent_clear_unlock_delalloc(inode, &BTRFS_I(inode)->io_tree, cur_offset, cur_offset + num_bytes - 1, locked_page, EXTENT_CLEAR_UNLOCK_PAGE | @@ -5904,16 +5911,15 @@ out_unlock: return err; } -static int prealloc_file_range(struct inode *inode, u64 start, u64 end, - u64 alloc_hint, int mode, loff_t actual_len) +int btrfs_prealloc_file_range(struct inode *inode, int mode, + u64 start, u64 num_bytes, u64 min_size, + loff_t actual_len, u64 *alloc_hint) { struct btrfs_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 cur_offset = start; - u64 num_bytes = end - start; int ret = 0; - u64 i_size; while (num_bytes > 0) { trans = btrfs_start_transaction(root, 3); @@ -5922,9 +5928,8 @@ static int prealloc_file_range(struct in break; } - ret = btrfs_reserve_extent(trans, root, num_bytes, - root->sectorsize, 0, alloc_hint, - (u64)-1, &ins, 1); + ret = btrfs_reserve_extent(trans, root, num_bytes, min_size, + 0, *alloc_hint, (u64)-1, &ins, 1); if (ret) { btrfs_end_transaction(trans, root); break; @@ -5941,20 +5946,19 @@ static int prealloc_file_range(struct in num_bytes -= ins.offset; cur_offset += ins.offset; - alloc_hint = ins.objectid + ins.offset; + *alloc_hint = ins.objectid + ins.offset; inode->i_ctime = CURRENT_TIME; BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC; if (!(mode & FALLOC_FL_KEEP_SIZE) && - (actual_len > inode->i_size) && - (cur_offset > inode->i_size)) { - + (actual_len > inode->i_size) && + (cur_offset > inode->i_size)) { if (cur_offset > actual_len) - i_size = actual_len; + i_size_write(inode, actual_len); else - i_size = cur_offset; - i_size_write(inode, i_size); - btrfs_ordered_update_i_size(inode, i_size, NULL); + i_size_write(inode, cur_offset); + i_size_write(inode, cur_offset); + btrfs_ordered_update_i_size(inode, cur_offset, NULL); } ret = btrfs_update_inode(trans, root, inode); @@ -6040,16 +6044,16 @@ static long btrfs_fallocate(struct inode if (em->block_start == EXTENT_MAP_HOLE || (cur_offset >= inode->i_size && !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) { - ret = prealloc_file_range(inode, - cur_offset, last_byte, - alloc_hint, mode, offset+len); + ret = btrfs_prealloc_file_range(inode, 0, cur_offset, + last_byte - cur_offset, + 1 << inode->i_blkbits, + offset + len, + &alloc_hint); if (ret < 0) { free_extent_map(em); break; } } - if (em->block_start <= EXTENT_MAP_LAST_BYTE) - alloc_hint = em->block_start; free_extent_map(em); cur_offset = last_byte; diff -urp 11/fs/btrfs/relocation.c 12/fs/btrfs/relocation.c --- 11/fs/btrfs/relocation.c 2010-04-18 10:24:51.294952000 +0800 +++ 12/fs/btrfs/relocation.c 2010-04-18 10:55:07.775699247 +0800 @@ -2545,6 +2545,50 @@ out: } static noinline_for_stack +int prealloc_file_extent_cluster(struct inode *inode, + struct file_extent_cluster *cluster) +{ + u64 alloc_hint = 0; + u64 start; + u64 end; + u64 offset = BTRFS_I(inode)->index_cnt; + u64 num_bytes; + int nr = 0; + int ret = 0; + + BUG_ON(cluster->start != cluster->boundary[0]); + mutex_lock(&inode->i_mutex); + + ret = btrfs_check_data_free_space(inode, cluster->end + + 1 - cluster->start); + if (ret) + goto out; + + while (nr < cluster->nr) { + start = cluster->boundary[nr] - offset; + if (nr + 1 < cluster->nr) + end = cluster->boundary[nr + 1] - 1 - offset; + else + end = cluster->end - offset; + + lock_extent(&BTRFS_I(inode)->io_tree, start, end, GFP_NOFS); + num_bytes = end + 1 - start; + ret = btrfs_prealloc_file_range(inode, 0, start, + num_bytes, num_bytes, + end + 1, &alloc_hint); + unlock_extent(&BTRFS_I(inode)->io_tree, start, end, GFP_NOFS); + if (ret) + break; + nr++; + } + btrfs_free_reserved_data_space(inode, cluster->end + + 1 - cluster->start); +out: + mutex_unlock(&inode->i_mutex); + return ret; +} + +static noinline_for_stack int setup_extent_mapping(struct inode *inode, u64 start, u64 end, u64 block_start) { @@ -2587,7 +2631,6 @@ static int relocate_file_extent_cluster( u64 offset = BTRFS_I(inode)->index_cnt; unsigned long index; unsigned long last_index; - unsigned int dirty_page = 0; struct page *page; struct file_ra_state *ra; int nr = 0; @@ -2600,21 +2643,24 @@ static int relocate_file_extent_cluster( if (!ra) return -ENOMEM; - index = (cluster->start - offset) >> PAGE_CACHE_SHIFT; - last_index = (cluster->end - offset) >> PAGE_CACHE_SHIFT; + ret = prealloc_file_extent_cluster(inode, cluster); + if (ret) + goto out; - mutex_lock(&inode->i_mutex); + file_ra_state_init(ra, inode->i_mapping); - i_size_write(inode, cluster->end + 1 - offset); ret = setup_extent_mapping(inode, cluster->start - offset, cluster->end - offset, cluster->start); if (ret) - goto out_unlock; - - file_ra_state_init(ra, inode->i_mapping); + goto out; - WARN_ON(cluster->start != cluster->boundary[0]); + index = (cluster->start - offset) >> PAGE_CACHE_SHIFT; + last_index = (cluster->end - offset) >> PAGE_CACHE_SHIFT; while (index <= last_index) { + ret = btrfs_delalloc_reserve_metadata(inode, PAGE_CACHE_SIZE); + if (ret) + goto out; + page = find_lock_page(inode->i_mapping, index); if (!page) { page_cache_sync_readahead(inode->i_mapping, @@ -2622,8 +2668,10 @@ static int relocate_file_extent_cluster( last_index + 1 - index); page = grab_cache_page(inode->i_mapping, index); if (!page) { + btrfs_delalloc_release_metadata(inode, + PAGE_CACHE_SIZE); ret = -ENOMEM; - goto out_unlock; + goto out; } } @@ -2639,8 +2687,10 @@ static int relocate_file_extent_cluster( if (!PageUptodate(page)) { unlock_page(page); page_cache_release(page); + btrfs_delalloc_release_metadata(inode, + PAGE_CACHE_SIZE); ret = -EIO; - goto out_unlock; + goto out; } } @@ -2659,10 +2709,9 @@ static int relocate_file_extent_cluster( EXTENT_BOUNDARY, GFP_NOFS); nr++; } - btrfs_set_extent_delalloc(inode, page_start, page_end, NULL); + btrfs_set_extent_delalloc(inode, page_start, page_end, NULL); set_page_dirty(page); - dirty_page++; unlock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end, GFP_NOFS); @@ -2670,20 +2719,11 @@ static int relocate_file_extent_cluster( page_cache_release(page); index++; - if (nr < cluster->nr && - page_end + 1 + offset == cluster->boundary[nr]) { - balance_dirty_pages_ratelimited_nr(inode->i_mapping, - dirty_page); - dirty_page = 0; - } - } - if (dirty_page) { - balance_dirty_pages_ratelimited_nr(inode->i_mapping, - dirty_page); + balance_dirty_pages_ratelimited(inode->i_mapping); + btrfs_throttle(BTRFS_I(inode)->root); } WARN_ON(nr != cluster->nr); -out_unlock: - mutex_unlock(&inode->i_mutex); +out: kfree(ra); return ret; }