From 34d52cb6c50b5a43901709998f59fb1c5a43dc4a Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 29 Mar 2011 13:46:06 +0800 Subject: Btrfs: Make free space cache code generic So we can re-use the code to cache free inode numbers. The change is quite straightforward. Two new structures are introduced. - struct btrfs_free_space_ctl We move those variables that are used for caching free space from struct btrfs_block_group_cache to this new struct. - struct btrfs_free_space_op We do block group specific work (e.g. calculation of extents threshold) through functions registered in this struct. And then we can remove references to struct btrfs_block_group_cache. Signed-off-by: Li Zefan --- fs/btrfs/ctree.h | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'fs/btrfs/ctree.h') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 2e61fe1b6b8..d17e4a3b8bf 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -830,9 +830,6 @@ struct btrfs_block_group_cache { u64 bytes_super; u64 flags; u64 sectorsize; - int extents_thresh; - int free_extents; - int total_bitmaps; unsigned int ro:1; unsigned int dirty:1; unsigned int iref:1; @@ -847,9 +844,7 @@ struct btrfs_block_group_cache { struct btrfs_space_info *space_info; /* free space cache stuff */ - spinlock_t tree_lock; - struct rb_root free_space_offset; - u64 free_space; + struct btrfs_free_space_ctl *free_space_ctl; /* block group cache stuff */ struct rb_node cache_node; -- cgit v1.2.3-18-g5258 From 581bb050941b4f220f84d3e5ed6dace3d42dd382 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 20 Apr 2011 10:06:11 +0800 Subject: Btrfs: Cache free inode numbers in memory Currently btrfs stores the highest objectid of the fs tree, and it always returns (highest+1) inode number when we create a file, so inode numbers won't be reclaimed when we delete files, so we'll run out of inode numbers as we keep create/delete files in 32bits machines. This fixes it, and it works similarly to how we cache free space in block cgroups. We start a kernel thread to read the file tree. By scanning inode items, we know which chunks of inode numbers are free, and we cache them in an rb-tree. Because we are searching the commit root, we have to carefully handle the cross-transaction case. The rb-tree is a hybrid extent+bitmap tree, so if we have too many small chunks of inode numbers, we'll use bitmaps. Initially we allow 16K ram of extents, and a bitmap will be used if we exceed this threshold. The extents threshold is adjusted in runtime. Signed-off-by: Li Zefan --- fs/btrfs/ctree.h | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'fs/btrfs/ctree.h') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index d17e4a3b8bf..c96a4e4c556 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1102,6 +1102,15 @@ struct btrfs_root { spinlock_t accounting_lock; struct btrfs_block_rsv *block_rsv; + /* free ino cache stuff */ + struct mutex fs_commit_mutex; + struct btrfs_free_space_ctl *free_ino_ctl; + enum btrfs_caching_type cached; + spinlock_t cache_lock; + wait_queue_head_t cache_wait; + struct btrfs_free_space_ctl *free_ino_pinned; + u64 cache_progress; + struct mutex log_mutex; wait_queue_head_t log_writer_wait; wait_queue_head_t log_commit_wait[2]; @@ -2408,12 +2417,6 @@ int btrfs_del_orphan_item(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 offset); int btrfs_find_orphan_item(struct btrfs_root *root, u64 offset); -/* inode-map.c */ -int btrfs_find_free_objectid(struct btrfs_trans_handle *trans, - struct btrfs_root *fs_root, - u64 dirid, u64 *objectid); -int btrfs_find_highest_inode(struct btrfs_root *fs_root, u64 *objectid); - /* inode-item.c */ int btrfs_insert_inode_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, -- cgit v1.2.3-18-g5258 From 82d5902d9c681be37ffa9d70482907f9f0b7ec1f Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 20 Apr 2011 10:33:24 +0800 Subject: Btrfs: Support reading/writing on disk free ino cache This is similar to block group caching. We dedicate a special inode in fs tree to save free ino cache. At the very first time we create/delete a file after mount, the free ino cache will be loaded from disk into memory. When the fs tree is commited, the cache will be written back to disk. To keep compatibility, we check the root generation against the generation of the special inode when loading the cache, so the loading will fail if the btrfs filesystem was mounted in an older kernel before. Signed-off-by: Li Zefan --- fs/btrfs/ctree.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'fs/btrfs/ctree.h') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index c96a4e4c556..b20082e27a9 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -105,6 +105,12 @@ struct btrfs_ordered_sum; /* For storing free space cache */ #define BTRFS_FREE_SPACE_OBJECTID -11ULL +/* + * The inode number assigned to the special inode for sotring + * free ino cache + */ +#define BTRFS_FREE_INO_OBJECTID -12ULL + /* dummy objectid represents multiple objectids */ #define BTRFS_MULTIPLE_OBJECTIDS -255ULL @@ -1110,6 +1116,7 @@ struct btrfs_root { wait_queue_head_t cache_wait; struct btrfs_free_space_ctl *free_ino_pinned; u64 cache_progress; + struct inode *cache_inode; struct mutex log_mutex; wait_queue_head_t log_writer_wait; -- cgit v1.2.3-18-g5258