From patchwork Wed Apr 6 18:56:39 2011 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Josef Bacik X-Patchwork-Id: 690351 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by demeter1.kernel.org (8.14.4/8.14.3) with ESMTP id p36J7VVg002948 for ; Wed, 6 Apr 2011 19:07:32 GMT Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756587Ab1DFTH3 (ORCPT ); Wed, 6 Apr 2011 15:07:29 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51045 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756538Ab1DFTH2 (ORCPT ); Wed, 6 Apr 2011 15:07:28 -0400 Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id p36J7SDi004903 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Wed, 6 Apr 2011 15:07:28 -0400 Received: from test1244.test.redhat.com (test1244.test.redhat.com [10.10.10.244]) by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id p36J7SMt005413 for ; Wed, 6 Apr 2011 15:07:28 -0400 From: Josef Bacik To: linux-btrfs@vger.kernel.org Subject: [PATCH] Btrfs: reuse the extent_map we found when calling btrfs_get_extent Date: Wed, 6 Apr 2011 14:56:39 -0400 Message-Id: <1302116199-3418-1-git-send-email-josef@redhat.com> X-Scanned-By: MIMEDefang 2.68 on 10.5.11.23 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-4.2.6 (demeter1.kernel.org [140.211.167.41]); Wed, 06 Apr 2011 19:07:32 +0000 (UTC) In btrfs_get_block_direct we call btrfs_get_extent to lookup the extent for the range that we are looking for. If we don't find an extent, btrfs_get_extent will insert a extent_map for that area and mark it as a hole. So it does the job of allocating a new extent map and inserting it into the io tree. But if we're creating a new extent we free it up and redo all of that work. So instead pass the em to btrfs_new_extent_direct(), and if it will work just allocate the disk space and set it up properly and bypass the freeing/allocating of a new extent map and the expensive operation of inserting the thing into the io_tree. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/inode.c | 36 ++++++++++++++++++++++++++++-------- 1 files changed, 28 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 93ed346..032e8fb 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -5443,17 +5443,30 @@ out: } static struct extent_map *btrfs_new_extent_direct(struct inode *inode, + struct extent_map *em, u64 start, u64 len) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; - struct extent_map *em; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; struct btrfs_key ins; u64 alloc_hint; int ret; + bool insert = false; - btrfs_drop_extent_cache(inode, start, start + len - 1, 0); + /* + * Ok if the extent map we looked up is a hole and is for the exact + * range we want, there is no reason to allocate a new one, however if + * it is not right then we need to free this one and drop the cache for + * our range. + */ + if (em->block_start != EXTENT_MAP_HOLE || em->start != start || + em->len != len) { + free_extent_map(em); + em = NULL; + insert = true; + btrfs_drop_extent_cache(inode, start, start + len - 1, 0); + } trans = btrfs_join_transaction(root, 0); if (IS_ERR(trans)) @@ -5469,10 +5482,12 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, goto out; } - em = alloc_extent_map(GFP_NOFS); if (!em) { - em = ERR_PTR(-ENOMEM); - goto out; + em = alloc_extent_map(GFP_NOFS); + if (!em) { + em = ERR_PTR(-ENOMEM); + goto out; + } } em->start = start; @@ -5482,9 +5497,15 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode, em->block_start = ins.objectid; em->block_len = ins.offset; em->bdev = root->fs_info->fs_devices->latest_bdev; + + /* + * We need to do this because if we're using the original em we searched + * for, we could have EXTENT_FLAG_VACANCY set, and we don't want that. + */ + em->flags = 0; set_bit(EXTENT_FLAG_PINNED, &em->flags); - while (1) { + while (insert) { write_lock(&em_tree->lock); ret = add_extent_mapping(em_tree, em); write_unlock(&em_tree->lock); @@ -5702,8 +5723,7 @@ must_cow: * it above */ len = bh_result->b_size; - free_extent_map(em); - em = btrfs_new_extent_direct(inode, start, len); + em = btrfs_new_extent_direct(inode, em, start, len); if (IS_ERR(em)) return PTR_ERR(em); len = min(len, em->len - (start - em->start));