diff options
Diffstat (limited to 'fs')
144 files changed, 3857 insertions, 1566 deletions
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 48d4215..c55c614 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -68,7 +68,7 @@ static int v9fs_set_super(struct super_block *s, void *data) * v9fs_fill_super - populate superblock with info * @sb: superblock * @v9ses: session information - * @flags: flags propagated from v9fs_get_sb() + * @flags: flags propagated from v9fs_mount() * */ @@ -99,18 +99,16 @@ v9fs_fill_super(struct super_block *sb, struct v9fs_session_info *v9ses, } /** - * v9fs_get_sb - mount a superblock + * v9fs_mount - mount a superblock * @fs_type: file system type * @flags: mount flags * @dev_name: device name that was mounted * @data: mount options - * @mnt: mountpoint record to be instantiated * */ -static int v9fs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { struct super_block *sb = NULL; struct inode *inode = NULL; @@ -124,7 +122,7 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags, v9ses = kzalloc(sizeof(struct v9fs_session_info), GFP_KERNEL); if (!v9ses) - return -ENOMEM; + return ERR_PTR(-ENOMEM); fid = v9fs_session_init(v9ses, dev_name, data); if (IS_ERR(fid)) { @@ -186,15 +184,15 @@ static int v9fs_get_sb(struct file_system_type *fs_type, int flags, v9fs_fid_add(root, fid); P9_DPRINTK(P9_DEBUG_VFS, " simple set mount, return 0\n"); - simple_set_mnt(mnt, sb); - return 0; + return dget(sb->s_root); clunk_fid: p9_client_clunk(fid); close_session: v9fs_session_close(v9ses); kfree(v9ses); - return retval; + return ERR_PTR(retval); + release_sb: /* * we will do the session_close and root dentry release @@ -204,7 +202,7 @@ release_sb: */ p9_client_clunk(fid); deactivate_locked_super(sb); - return retval; + return ERR_PTR(retval); } /** @@ -300,7 +298,7 @@ static const struct super_operations v9fs_super_ops_dotl = { struct file_system_type v9fs_fs_type = { .name = "9p", - .get_sb = v9fs_get_sb, + .mount = v9fs_mount, .kill_sb = v9fs_kill_super, .owner = THIS_MODULE, .fs_flags = FS_RENAME_DOES_D_MOVE, diff --git a/fs/adfs/super.c b/fs/adfs/super.c index d9803f7..959dbff 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -490,17 +490,16 @@ error: return -EINVAL; } -static int adfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *adfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, adfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, adfs_fill_super); } static struct file_system_type adfs_fs_type = { .owner = THIS_MODULE, .name = "adfs", - .get_sb = adfs_get_sb, + .mount = adfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/affs/super.c b/fs/affs/super.c index fa4fbe1..0cf7f43 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -573,17 +573,16 @@ affs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int affs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *affs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, affs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, affs_fill_super); } static struct file_system_type affs_fs_type = { .owner = THIS_MODULE, .name = "affs", - .get_sb = affs_get_sb, + .mount = affs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/afs/super.c b/fs/afs/super.c index eacf76d..27201cff 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -29,9 +29,8 @@ #define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */ static void afs_i_init_once(void *foo); -static int afs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data, struct vfsmount *mnt); +static struct dentry *afs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data); static struct inode *afs_alloc_inode(struct super_block *sb); static void afs_put_super(struct super_block *sb); static void afs_destroy_inode(struct inode *inode); @@ -40,7 +39,7 @@ static int afs_statfs(struct dentry *dentry, struct kstatfs *buf); struct file_system_type afs_fs_type = { .owner = THIS_MODULE, .name = "afs", - .get_sb = afs_get_sb, + .mount = afs_mount, .kill_sb = kill_anon_super, .fs_flags = 0, }; @@ -359,11 +358,8 @@ error: /* * get an AFS superblock */ -static int afs_get_sb(struct file_system_type *fs_type, - int flags, - const char *dev_name, - void *options, - struct vfsmount *mnt) +static struct dentry *afs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *options) { struct afs_mount_params params; struct super_block *sb; @@ -427,12 +423,11 @@ static int afs_get_sb(struct file_system_type *fs_type, ASSERTCMP(sb->s_flags, &, MS_ACTIVE); } - simple_set_mnt(mnt, sb); afs_put_volume(params.volume); afs_put_cell(params.cell); kfree(new_opts); _leave(" = 0 [%p]", sb); - return 0; + return dget(sb->s_root); error: afs_put_volume(params.volume); @@ -440,7 +435,7 @@ error: key_put(params.key); kfree(new_opts); _leave(" = %d", ret); - return ret; + return ERR_PTR(ret); } /* diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 5365527..57ce55b 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -26,12 +26,10 @@ static struct vfsmount *anon_inode_mnt __read_mostly; static struct inode *anon_inode_inode; static const struct file_operations anon_inode_fops; -static int anon_inodefs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *anon_inodefs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_pseudo(fs_type, "anon_inode:", NULL, ANON_INODE_FS_MAGIC, - mnt); + return mount_pseudo(fs_type, "anon_inode:", NULL, ANON_INODE_FS_MAGIC); } /* @@ -45,7 +43,7 @@ static char *anon_inodefs_dname(struct dentry *dentry, char *buffer, int buflen) static struct file_system_type anon_inode_fs_type = { .name = "anon_inodefs", - .get_sb = anon_inodefs_get_sb, + .mount = anon_inodefs_mount, .kill_sb = kill_anon_super, }; static const struct dentry_operations anon_inodefs_dentry_operations = { diff --git a/fs/autofs4/init.c b/fs/autofs4/init.c index 9722e4b..c038727 100644 --- a/fs/autofs4/init.c +++ b/fs/autofs4/init.c @@ -14,16 +14,16 @@ #include <linux/init.h> #include "autofs_i.h" -static int autofs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *autofs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, autofs4_fill_super, mnt); + return mount_nodev(fs_type, flags, data, autofs4_fill_super); } static struct file_system_type autofs_fs_type = { .owner = THIS_MODULE, .name = "autofs", - .get_sb = autofs_get_sb, + .mount = autofs_mount, .kill_sb = autofs4_kill_sb, }; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index dc39d28..aa4e7c7 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -913,18 +913,17 @@ befs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int -befs_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) +static struct dentry * +befs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, befs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, befs_fill_super); } static struct file_system_type befs_fs_type = { .owner = THIS_MODULE, .name = "befs", - .get_sb = befs_get_sb, + .mount = befs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 883e77a..76db6d7 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -450,16 +450,16 @@ out: return ret; } -static int bfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *bfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, bfs_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, bfs_fill_super); } static struct file_system_type bfs_fs_type = { .owner = THIS_MODULE, .name = "bfs", - .get_sb = bfs_get_sb, + .mount = bfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 29990f0..1befe2e 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -706,10 +706,10 @@ static int bm_fill_super(struct super_block * sb, void * data, int silent) return err; } -static int bm_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *bm_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_single(fs_type, flags, data, bm_fill_super, mnt); + return mount_single(fs_type, flags, data, bm_fill_super); } static struct linux_binfmt misc_format = { @@ -720,7 +720,7 @@ static struct linux_binfmt misc_format = { static struct file_system_type bm_fs_type = { .owner = THIS_MODULE, .name = "binfmt_misc", - .get_sb = bm_get_sb, + .mount = bm_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/block_dev.c b/fs/block_dev.c index dea3b62..06e8ff1 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -464,15 +464,15 @@ static const struct super_operations bdev_sops = { .evict_inode = bdev_evict_inode, }; -static int bd_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *bd_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_pseudo(fs_type, "bdev:", &bdev_sops, 0x62646576, mnt); + return mount_pseudo(fs_type, "bdev:", &bdev_sops, 0x62646576); } static struct file_system_type bd_type = { .name = "bdev", - .get_sb = bd_get_sb, + .mount = bd_mount, .kill_sb = kill_anon_super, }; diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 396039b..7845d1f 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -163,7 +163,6 @@ fail: */ static void end_compressed_bio_read(struct bio *bio, int err) { - struct extent_io_tree *tree; struct compressed_bio *cb = bio->bi_private; struct inode *inode; struct page *page; @@ -187,7 +186,6 @@ static void end_compressed_bio_read(struct bio *bio, int err) /* ok, we're the last bio for this extent, lets start * the decompression. */ - tree = &BTRFS_I(inode)->io_tree; ret = btrfs_zlib_decompress_biovec(cb->compressed_pages, cb->start, cb->orig_bio->bi_io_vec, diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index c3df14c..9ac1715 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -200,7 +200,6 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans, struct extent_buffer **cow_ret, u64 new_root_objectid) { struct extent_buffer *cow; - u32 nritems; int ret = 0; int level; struct btrfs_disk_key disk_key; @@ -210,7 +209,6 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans, WARN_ON(root->ref_cows && trans->transid != root->last_trans); level = btrfs_header_level(buf); - nritems = btrfs_header_nritems(buf); if (level == 0) btrfs_item_key(buf, &disk_key, 0); else @@ -1008,7 +1006,6 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, int wret; int pslot; int orig_slot = path->slots[level]; - int err_on_enospc = 0; u64 orig_ptr; if (level == 0) @@ -1071,8 +1068,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, BTRFS_NODEPTRS_PER_BLOCK(root) / 4) return 0; - if (btrfs_header_nritems(mid) < 2) - err_on_enospc = 1; + btrfs_header_nritems(mid); left = read_node_slot(root, parent, pslot - 1); if (left) { @@ -1103,8 +1099,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, wret = push_node_left(trans, root, left, mid, 1); if (wret < 0) ret = wret; - if (btrfs_header_nritems(mid) < 2) - err_on_enospc = 1; + btrfs_header_nritems(mid); } /* @@ -1224,14 +1219,12 @@ static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans, int wret; int pslot; int orig_slot = path->slots[level]; - u64 orig_ptr; if (level == 0) return 1; mid = path->nodes[level]; WARN_ON(btrfs_header_generation(mid) != trans->transid); - orig_ptr = btrfs_node_blockptr(mid, orig_slot); if (level < BTRFS_MAX_LEVEL - 1) parent = path->nodes[level + 1]; @@ -1577,13 +1570,33 @@ read_block_for_search(struct btrfs_trans_handle *trans, blocksize = btrfs_level_size(root, level - 1); tmp = btrfs_find_tree_block(root, blocknr, blocksize); - if (tmp && btrfs_buffer_uptodate(tmp, gen)) { - /* - * we found an up to date block without sleeping, return - * right away - */ - *eb_ret = tmp; - return 0; + if (tmp) { + if (btrfs_buffer_uptodate(tmp, 0)) { + if (btrfs_buffer_uptodate(tmp, gen)) { + /* + * we found an up to date block without + * sleeping, return + * right away + */ + *eb_ret = tmp; + return 0; + } + /* the pages were up to date, but we failed + * the generation number check. Do a full + * read for the generation number that is correct. + * We must do this without dropping locks so + * we can trust our generation number + */ + free_extent_buffer(tmp); + tmp = read_tree_block(root, blocknr, blocksize, gen); + if (tmp && btrfs_buffer_uptodate(tmp, gen)) { + *eb_ret = tmp; + return 0; + } + free_extent_buffer(tmp); + btrfs_release_path(NULL, p); + return -EIO; + } } /* @@ -1596,8 +1609,7 @@ read_block_for_search(struct btrfs_trans_handle *trans, btrfs_unlock_up_safe(p, level + 1); btrfs_set_path_blocking(p); - if (tmp) - free_extent_buffer(tmp); + free_extent_buffer(tmp); if (p->reada) reada_for_search(root, p, level, slot, key->objectid); @@ -2548,7 +2560,6 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, { struct btrfs_disk_key disk_key; struct extent_buffer *right = path->nodes[0]; - int slot; int i; int push_space = 0; int push_items = 0; @@ -2560,8 +2571,6 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, u32 this_item_size; u32 old_left_item_size; - slot = path->slots[1]; - if (empty) nr = min(right_nritems, max_slot); else @@ -3330,7 +3339,6 @@ int btrfs_truncate_item(struct btrfs_trans_handle *trans, { int ret = 0; int slot; - int slot_orig; struct extent_buffer *leaf; struct btrfs_item *item; u32 nritems; @@ -3340,7 +3348,6 @@ int btrfs_truncate_item(struct btrfs_trans_handle *trans, unsigned int size_diff; int i; - slot_orig = path->slots[0]; leaf = path->nodes[0]; slot = path->slots[0]; @@ -3445,7 +3452,6 @@ int btrfs_extend_item(struct btrfs_trans_handle *trans, { int ret = 0; int slot; - int slot_orig; struct extent_buffer *leaf; struct btrfs_item *item; u32 nritems; @@ -3454,7 +3460,6 @@ int btrfs_extend_item(struct btrfs_trans_handle *trans, unsigned int old_size; int i; - slot_orig = path->slots[0]; leaf = path->nodes[0]; nritems = btrfs_header_nritems(leaf); @@ -3787,7 +3792,6 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans, struct btrfs_key *cpu_key, u32 *data_size, int nr) { - struct extent_buffer *leaf; int ret = 0; int slot; int i; @@ -3804,7 +3808,6 @@ int btrfs_insert_empty_items(struct btrfs_trans_handle *trans, if (ret < 0) goto out; - leaf = path->nodes[0]; slot = path->slots[0]; BUG_ON(slot < 0); diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index eaf286a..8db9234 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -99,6 +99,9 @@ struct btrfs_ordered_sum; */ #define BTRFS_EXTENT_CSUM_OBJECTID -10ULL +/* For storing free space cache */ +#define BTRFS_FREE_SPACE_OBJECTID -11ULL + /* dummy objectid represents multiple objectids */ #define BTRFS_MULTIPLE_OBJECTIDS -255ULL @@ -265,6 +268,22 @@ struct btrfs_chunk { /* additional stripes go here */ } __attribute__ ((__packed__)); +#define BTRFS_FREE_SPACE_EXTENT 1 +#define BTRFS_FREE_SPACE_BITMAP 2 + +struct btrfs_free_space_entry { + __le64 offset; + __le64 bytes; + u8 type; +} __attribute__ ((__packed__)); + +struct btrfs_free_space_header { + struct btrfs_disk_key location; + __le64 generation; + __le64 num_entries; + __le64 num_bitmaps; +} __attribute__ ((__packed__)); + static inline unsigned long btrfs_chunk_item_size(int num_stripes) { BUG_ON(num_stripes == 0); @@ -365,8 +384,10 @@ struct btrfs_super_block { char label[BTRFS_LABEL_SIZE]; + __le64 cache_generation; + /* future expansion */ - __le64 reserved[32]; + __le64 reserved[31]; u8 sys_chunk_array[BTRFS_SYSTEM_CHUNK_ARRAY_SIZE]; } __attribute__ ((__packed__)); @@ -375,13 +396,15 @@ struct btrfs_super_block { * ones specified below then we will fail to mount */ #define BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF (1ULL << 0) -#define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL (2ULL << 0) +#define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL (1ULL << 1) +#define BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS (1ULL << 2) #define BTRFS_FEATURE_COMPAT_SUPP 0ULL #define BTRFS_FEATURE_COMPAT_RO_SUPP 0ULL -#define BTRFS_FEATURE_INCOMPAT_SUPP \ - (BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF | \ - BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL) +#define BTRFS_FEATURE_INCOMPAT_SUPP \ + (BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF | \ + BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL | \ + BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS) /* * A leaf is full of items. offset and size tell us where to find @@ -675,7 +698,8 @@ struct btrfs_block_group_item { struct btrfs_space_info { u64 flags; - u64 total_bytes; /* total bytes in the space */ + u64 total_bytes; /* total bytes in the space, + this doesn't take mirrors into account */ u64 bytes_used; /* total bytes used, this does't take mirrors into account */ u64 bytes_pinned; /* total bytes pinned, will be freed when the @@ -687,6 +711,8 @@ struct btrfs_space_info { u64 bytes_may_use; /* number of bytes that may be used for delalloc/allocations */ u64 disk_used; /* total bytes used on disk */ + u64 disk_total; /* total bytes on disk, takes mirrors into + account */ int full; /* indicates that we cannot allocate any more chunks for this space */ @@ -750,6 +776,14 @@ enum btrfs_caching_type { BTRFS_CACHE_FINISHED = 2, }; +enum btrfs_disk_cache_state { + BTRFS_DC_WRITTEN = 0, + BTRFS_DC_ERROR = 1, + BTRFS_DC_CLEAR = 2, + BTRFS_DC_SETUP = 3, + BTRFS_DC_NEED_WRITE = 4, +}; + struct btrfs_caching_control { struct list_head list; struct mutex mutex; @@ -763,6 +797,7 @@ struct btrfs_block_group_cache { struct btrfs_key key; struct btrfs_block_group_item item; struct btrfs_fs_info *fs_info; + struct inode *inode; spinlock_t lock; u64 pinned; u64 reserved; @@ -773,8 +808,11 @@ struct btrfs_block_group_cache { int extents_thresh; int free_extents; int total_bitmaps; - int ro; - int dirty; + int ro:1; + int dirty:1; + int iref:1; + + int disk_cache_state; /* cache tracking stuff */ int cached; @@ -863,6 +901,7 @@ struct btrfs_fs_info { struct btrfs_transaction *running_transaction; wait_queue_head_t transaction_throttle; wait_queue_head_t transaction_wait; + wait_queue_head_t transaction_blocked_wait; wait_queue_head_t async_submit_wait; struct btrfs_super_block super_copy; @@ -949,6 +988,7 @@ struct btrfs_fs_info { struct btrfs_workers endio_meta_workers; struct btrfs_workers endio_meta_write_workers; struct btrfs_workers endio_write_workers; + struct btrfs_workers endio_freespace_worker; struct btrfs_workers submit_workers; /* * fixup workers take dirty pages that didn't properly go through @@ -1192,6 +1232,9 @@ struct btrfs_root { #define BTRFS_MOUNT_NOSSD (1 << 9) #define BTRFS_MOUNT_DISCARD (1 << 10) #define BTRFS_MOUNT_FORCE_COMPRESS (1 << 11) +#define BTRFS_MOUNT_SPACE_CACHE (1 << 12) +#define BTRFS_MOUNT_CLEAR_CACHE (1 << 13) +#define BTRFS_MOUNT_USER_SUBVOL_RM_ALLOWED (1 << 14) #define btrfs_clear_opt(o, opt) ((o) &= ~BTRFS_MOUNT_##opt) #define btrfs_set_opt(o, opt) ((o) |= BTRFS_MOUNT_##opt) @@ -1665,6 +1708,27 @@ static inline void btrfs_set_dir_item_key(struct extent_buffer *eb, write_eb_member(eb, item, struct btrfs_dir_item, location, key); } +BTRFS_SETGET_FUNCS(free_space_entries, struct btrfs_free_space_header, + num_entries, 64); +BTRFS_SETGET_FUNCS(free_space_bitmaps, struct btrfs_free_space_header, + num_bitmaps, 64); +BTRFS_SETGET_FUNCS(free_space_generation, struct btrfs_free_space_header, + generation, 64); + +static inline void btrfs_free_space_key(struct extent_buffer *eb, + struct btrfs_free_space_header *h, + struct btrfs_disk_key *key) +{ + read_eb_member(eb, h, struct btrfs_free_space_header, location, key); +} + +static inline void btrfs_set_free_space_key(struct extent_buffer *eb, + struct btrfs_free_space_header *h, + struct btrfs_disk_key *key) +{ + write_eb_member(eb, h, struct btrfs_free_space_header, location, key); +} + /* struct btrfs_disk_key */ BTRFS_SETGET_STACK_FUNCS(disk_key_objectid, struct btrfs_disk_key, objectid, 64); @@ -1876,6 +1940,8 @@ BTRFS_SETGET_STACK_FUNCS(super_incompat_flags, struct btrfs_super_block, incompat_flags, 64); BTRFS_SETGET_STACK_FUNCS(super_csum_type, struct btrfs_super_block, csum_type, 16); +BTRFS_SETGET_STACK_FUNCS(super_cache_generation, struct btrfs_super_block, + cache_generation, 64); static inline int btrfs_super_csum_size(struct btrfs_super_block *s) { @@ -1988,6 +2054,12 @@ static inline struct dentry *fdentry(struct file *file) return file->f_path.dentry; } +static inline bool btrfs_mixed_space_info(struct btrfs_space_info *space_info) +{ + return ((space_info->flags & BTRFS_BLOCK_GROUP_METADATA) && + (space_info->flags & BTRFS_BLOCK_GROUP_DATA)); +} + /* extent-tree.c */ void btrfs_put_block_group(struct btrfs_block_group_cache *cache); int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, @@ -2079,7 +2151,7 @@ int btrfs_check_data_free_space(struct inode *inode, u64 bytes); void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes); int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans, struct btrfs_root *root, - int num_items, int *retries); + int num_items); void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans, struct btrfs_root *root); int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans, @@ -2100,7 +2172,7 @@ void btrfs_add_durable_block_rsv(struct btrfs_fs_info *fs_info, int btrfs_block_rsv_add(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, - u64 num_bytes, int *retries); + u64 num_bytes); int btrfs_block_rsv_check(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, @@ -2115,6 +2187,7 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, struct btrfs_block_group_cache *cache); int btrfs_set_block_group_rw(struct btrfs_root *root, struct btrfs_block_group_cache *cache); +void btrfs_put_block_group_cache(struct btrfs_fs_info *info); /* ctree.c */ int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key, int level, int *slot); @@ -2373,7 +2446,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, u32 min_type); int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput); -int btrfs_start_one_delalloc_inode(struct btrfs_root *root, int delay_iput); +int btrfs_start_one_delalloc_inode(struct btrfs_root *root, int delay_iput, + int sync); int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end, struct extent_state **cached_state); int btrfs_writepages(struct address_space *mapping, @@ -2426,6 +2500,10 @@ 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); +int btrfs_prealloc_file_range_trans(struct inode *inode, + struct btrfs_trans_handle *trans, 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 --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index e9103b3..f0cad5a 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -427,5 +427,5 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans, ret = btrfs_truncate_item(trans, root, path, item_len - sub_item_len, 1); } - return 0; + return ret; } diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 5e789f4..fb827d0 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -338,7 +338,6 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page) struct extent_io_tree *tree; u64 start = (u64)page->index << PAGE_CACHE_SHIFT; u64 found_start; - int found_level; unsigned long len; struct extent_buffer *eb; int ret; @@ -369,8 +368,6 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page) WARN_ON(1); goto err; } - found_level = btrfs_header_level(eb); - csum_tree_block(root, eb, 0); err: free_extent_buffer(eb); @@ -481,9 +478,12 @@ static void end_workqueue_bio(struct bio *bio, int err) end_io_wq->work.flags = 0; if (bio->bi_rw & REQ_WRITE) { - if (end_io_wq->metadata) + if (end_io_wq->metadata == 1) btrfs_queue_worker(&fs_info->endio_meta_write_workers, &end_io_wq->work); + else if (end_io_wq->metadata == 2) + btrfs_queue_worker(&fs_info->endio_freespace_worker, + &end_io_wq->work); else btrfs_queue_worker(&fs_info->endio_write_workers, &end_io_wq->work); @@ -497,6 +497,13 @@ static void end_workqueue_bio(struct bio *bio, int err) } } +/* + * For the metadata arg you want + * + * 0 - if data + * 1 - if normal metadta + * 2 - if writing to the free space cache area + */ int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio, int metadata) { @@ -533,11 +540,9 @@ int btrfs_congested_async(struct btrfs_fs_info *info, int iodone) static void run_one_async_start(struct btrfs_work *work) { - struct btrfs_fs_info *fs_info; struct async_submit_bio *async; async = container_of(work, struct async_submit_bio, work); - fs_info = BTRFS_I(async->inode)->root->fs_info; async->submit_bio_start(async->inode, async->rw, async->bio, async->mirror_num, async->bio_flags, async->bio_offset); @@ -850,12 +855,8 @@ struct extent_buffer *read_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize, u64 parent_transid) { struct extent_buffer *buf = NULL; - struct inode *btree_inode = root->fs_info->btree_inode; - struct extent_io_tree *io_tree; int ret; - io_tree = &BTRFS_I(btree_inode)->io_tree; - buf = btrfs_find_create_tree_block(root, bytenr, blocksize); if (!buf) return NULL; @@ -1377,7 +1378,6 @@ static int bio_ready_for_csum(struct bio *bio) u64 start = 0; struct page *page; struct extent_io_tree *io_tree = NULL; - struct btrfs_fs_info *info = NULL; struct bio_vec *bvec; int i; int ret; @@ -1396,7 +1396,6 @@ static int bio_ready_for_csum(struct bio *bio) buf_len = page->private >> 2; start = page_offset(page) + bvec->bv_offset; io_tree = &BTRFS_I(page->mapping->host)->io_tree; - info = BTRFS_I(page->mapping->host)->root->fs_info; } /* are we fully contained in this bio? */ if (buf_len <= length) @@ -1680,12 +1679,12 @@ struct btrfs_root *open_ctree(struct super_block *sb, init_waitqueue_head(&fs_info->transaction_throttle); init_waitqueue_head(&fs_info->transaction_wait); + init_waitqueue_head(&fs_info->transaction_blocked_wait); init_waitqueue_head(&fs_info->async_submit_wait); __setup_root(4096, 4096, 4096, 4096, tree_root, fs_info, BTRFS_ROOT_TREE_OBJECTID); - bh = btrfs_read_dev_super(fs_devices->latest_bdev); if (!bh) goto fail_iput; @@ -1775,6 +1774,8 @@ struct btrfs_root *open_ctree(struct super_block *sb, btrfs_init_workers(&fs_info->endio_write_workers, "endio-write", fs_info->thread_pool_size, &fs_info->generic_worker); + btrfs_init_workers(&fs_info->endio_freespace_worker, "freespace-write", + 1, &fs_info->generic_worker); /* * endios are largely parallel and should have a very @@ -1795,6 +1796,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, btrfs_start_workers(&fs_info->endio_meta_workers, 1); btrfs_start_workers(&fs_info->endio_meta_write_workers, 1); btrfs_start_workers(&fs_info->endio_write_workers, 1); + btrfs_start_workers(&fs_info->endio_freespace_worker, 1); fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super); fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages, @@ -1993,6 +1995,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, if (!(sb->s_flags & MS_RDONLY)) { down_read(&fs_info->cleanup_work_sem); btrfs_orphan_cleanup(fs_info->fs_root); + btrfs_orphan_cleanup(fs_info->tree_root); up_read(&fs_info->cleanup_work_sem); } @@ -2035,6 +2038,7 @@ fail_sb_buffer: btrfs_stop_workers(&fs_info->endio_meta_workers); btrfs_stop_workers(&fs_info->endio_meta_write_workers); btrfs_stop_workers(&fs_info->endio_write_workers); + btrfs_stop_workers(&fs_info->endio_freespace_worker); btrfs_stop_workers(&fs_info->submit_workers); fail_iput: invalidate_inode_pages2(fs_info->btree_inode->i_mapping); @@ -2410,6 +2414,7 @@ int close_ctree(struct btrfs_root *root) fs_info->closing = 1; smp_mb(); + btrfs_put_block_group_cache(fs_info); if (!(fs_info->sb->s_flags & MS_RDONLY)) { ret = btrfs_commit_super(root); if (ret) @@ -2456,6 +2461,7 @@ int close_ctree(struct btrfs_root *root) btrfs_stop_workers(&fs_info->endio_meta_workers); btrfs_stop_workers(&fs_info->endio_meta_write_workers); btrfs_stop_workers(&fs_info->endio_write_workers); + btrfs_stop_workers(&fs_info->endio_freespace_worker); btrfs_stop_workers(&fs_info->submit_workers); btrfs_close_devices(fs_info->fs_devices); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0b81ecd..0c097f3 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -242,6 +242,12 @@ get_caching_control(struct btrfs_block_group_cache *cache) return NULL; } + /* We're loading it the fast way, so we don't have a caching_ctl. */ + if (!cache->caching_ctl) { + spin_unlock(&cache->lock); + return NULL; + } + ctl = cache->caching_ctl; atomic_inc(&ctl->count); spin_unlock(&cache->lock); @@ -421,7 +427,9 @@ err: return 0; } -static int cache_block_group(struct btrfs_block_group_cache *cache) +static int cache_block_group(struct btrfs_block_group_cache *cache, + struct btrfs_trans_handle *trans, + int load_cache_only) { struct btrfs_fs_info *fs_info = cache->fs_info; struct btrfs_caching_control *caching_ctl; @@ -432,6 +440,36 @@ static int cache_block_group(struct btrfs_block_group_cache *cache) if (cache->cached != BTRFS_CACHE_NO) return 0; + /* + * We can't do the read from on-disk cache during a commit since we need + * to have the normal tree locking. + */ + if (!trans->transaction->in_commit) { + spin_lock(&cache->lock); + if (cache->cached != BTRFS_CACHE_NO) { + spin_unlock(&cache->lock); + return 0; + } + cache->cached = BTRFS_CACHE_STARTED; + spin_unlock(&cache->lock); + + ret = load_free_space_cache(fs_info, cache); + + spin_lock(&cache->lock); + if (ret == 1) { + cache->cached = BTRFS_CACHE_FINISHED; + cache->last_byte_to_unpin = (u64)-1; + } else { + cache->cached = BTRFS_CACHE_NO; + } + spin_unlock(&cache->lock); + if (ret == 1) + return 0; + } + + if (load_cache_only) + return 0; + caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_KERNEL); BUG_ON(!caching_ctl); @@ -509,7 +547,7 @@ static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info, rcu_read_lock(); list_for_each_entry_rcu(found, head, list) { - if (found->flags == flags) { + if (found->flags & flags) { rcu_read_unlock(); return found; } @@ -542,6 +580,15 @@ static u64 div_factor(u64 num, int factor) return num; } +static u64 div_factor_fine(u64 num, int factor) +{ + if (factor == 100) + return num; + num *= factor; + do_div(num, 100); + return num; +} + u64 btrfs_find_block_group(struct btrfs_root *root, u64 search_start, u64 search_hint, int owner) { @@ -2687,6 +2734,109 @@ next_block_group(struct btrfs_root *root, return cache; } +static int cache_save_setup(struct btrfs_block_group_cache *block_group, + struct btrfs_trans_handle *trans, + struct btrfs_path *path) +{ + struct btrfs_root *root = block_group->fs_info->tree_root; + struct inode *inode = NULL; + u64 alloc_hint = 0; + int num_pages = 0; + int retries = 0; + int ret = 0; + + /* + * If this block group is smaller than 100 megs don't bother caching the + * block group. + */ + if (block_group->key.offset < (100 * 1024 * 1024)) { + spin_lock(&block_group->lock); + block_group->disk_cache_state = BTRFS_DC_WRITTEN; + spin_unlock(&block_group->lock); + return 0; + } + +again: + inode = lookup_free_space_inode(root, block_group, path); + if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) { + ret = PTR_ERR(inode); + btrfs_release_path(root, path); + goto out; + } + + if (IS_ERR(inode)) { + BUG_ON(retries); + retries++; + + if (block_group->ro) + goto out_free; + + ret = create_free_space_inode(root, trans, block_group, path); + if (ret) + goto out_free; + goto again; + } + + /* + * We want to set the generation to 0, that way if anything goes wrong + * from here on out we know not to trust this cache when we load up next + * time. + */ + BTRFS_I(inode)->generation = 0; + ret = btrfs_update_inode(trans, root, inode); + WARN_ON(ret); + + if (i_size_read(inode) > 0) { + ret = btrfs_truncate_free_space_cache(root, trans, path, + inode); + if (ret) + goto out_put; + } + + spin_lock(&block_group->lock); + if (block_group->cached != BTRFS_CACHE_FINISHED) { + spin_unlock(&block_group->lock); + goto out_put; + } + spin_unlock(&block_group->lock); + + num_pages = (int)div64_u64(block_group->key.offset, 1024 * 1024 * 1024); + if (!num_pages) + num_pages = 1; + + /* + * Just to make absolutely sure we have enough space, we're going to + * preallocate 12 pages worth of space for each block group. In + * practice we ought to use at most 8, but we need extra space so we can + * add our header and have a terminator between the extents and the + * bitmaps. + */ + num_pages *= 16; + num_pages *= PAGE_CACHE_SIZE; + + ret = btrfs_check_data_free_space(inode, num_pages); + if (ret) + goto out_put; + + ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, num_pages, + num_pages, num_pages, + &alloc_hint); + btrfs_free_reserved_data_space(inode, num_pages); +out_put: + iput(inode); +out_free: + btrfs_release_path(root, path); +out: + spin_lock(&block_group->lock); + if (ret) + block_group->disk_cache_state = BTRFS_DC_ERROR; + else + block_group->disk_cache_state = BTRFS_DC_SETUP; + spin_unlock(&block_group->lock); + + return ret; +} + int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans, struct btrfs_root *root) { @@ -2699,6 +2849,25 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; +again: + while (1) { + cache = btrfs_lookup_first_block_group(root->fs_info, last); + while (cache) { + if (cache->disk_cache_state == BTRFS_DC_CLEAR) + break; + cache = next_block_group(root, cache); + } + if (!cache) { + if (last == 0) + break; + last = 0; + continue; + } + err = cache_save_setup(cache, trans, path); + last = cache->key.objectid + cache->key.offset; + btrfs_put_block_group(cache); + } + while (1) { if (last == 0) { err = btrfs_run_delayed_refs(trans, root, @@ -2708,6 +2877,11 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans, cache = btrfs_lookup_first_block_group(root->fs_info, last); while (cache) { + if (cache->disk_cache_state == BTRFS_DC_CLEAR) { + btrfs_put_block_group(cache); + goto again; + } + if (cache->dirty) break; cache = next_block_group(root, cache); @@ -2719,6 +2893,8 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans, continue; } + if (cache->disk_cache_state == BTRFS_DC_SETUP) + cache->disk_cache_state = BTRFS_DC_NEED_WRITE; cache->dirty = 0; last = cache->key.objectid + cache->key.offset; @@ -2727,6 +2903,52 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans, btrfs_put_block_group(cache); } + while (1) { + /* + * I don't think this is needed since we're just marking our + * preallocated extent as written, but just in case it can't + * hurt. + */ + if (last == 0) { + err = btrfs_run_delayed_refs(trans, root, + (unsigned long)-1); + BUG_ON(err); + } + + cache = btrfs_lookup_first_block_group(root->fs_info, last); + while (cache) { + /* + * Really this shouldn't happen, but it could if we + * couldn't write the entire preallocated extent and + * splitting the extent resulted in a new block. + */ + if (cache->dirty) { + btrfs_put_block_group(cache); + goto again; + } + if (cache->disk_cache_state == BTRFS_DC_NEED_WRITE) + break; + cache = next_block_group(root, cache); + } + if (!cache) { + if (last == 0) + break; + last = 0; + continue; + } + + btrfs_write_out_cache(root, trans, cache, path); + + /* + * If we didn't have an error then the cache state is still + * NEED_WRITE, so we can set it to WRITTEN. + */ + if (cache->disk_cache_state == BTRFS_DC_NEED_WRITE) + cache->disk_cache_state = BTRFS_DC_WRITTEN; + last = cache->key.objectid + cache->key.offset; + btrfs_put_block_group(cache); + } + btrfs_free_path(path); return 0; } @@ -2762,6 +2984,7 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags, if (found) { spin_lock(&found->lock); found->total_bytes += total_bytes; + found->disk_total += total_bytes * factor; found->bytes_used += bytes_used; found->disk_used += bytes_used * factor; found->full = 0; @@ -2781,6 +3004,7 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags, BTRFS_BLOCK_GROUP_SYSTEM | BTRFS_BLOCK_GROUP_METADATA); found->total_bytes = total_bytes; + found->disk_total = total_bytes * factor; found->bytes_used = bytes_used; found->disk_used = bytes_used * factor; found->bytes_pinned = 0; @@ -2882,11 +3106,16 @@ int btrfs_check_data_free_space(struct inode *inode, u64 bytes) struct btrfs_space_info *data_sinfo; struct btrfs_root *root = BTRFS_I(inode)->root; u64 used; - int ret = 0, committed = 0; + int ret = 0, committed = 0, alloc_chunk = 1; /* make sure bytes are sectorsize aligned */ bytes = (bytes + root->sectorsize - 1) & ~((u64)root->sectorsize - 1); + if (root == root->fs_info->tree_root) { + alloc_chunk = 0; + committed = 1; + } + data_sinfo = BTRFS_I(inode)->space_info; if (!data_sinfo) goto alloc; @@ -2905,7 +3134,7 @@ again: * if we don't have enough free bytes in this space then we need * to alloc a new chunk. */ - if (!data_sinfo->full) { + if (!data_sinfo->full && alloc_chunk) { u64 alloc_target; data_sinfo->force_alloc = 1; @@ -2997,10 +3226,11 @@ static void force_metadata_allocation(struct btrfs_fs_info *info) rcu_read_unlock(); } -static int should_alloc_chunk(struct btrfs_space_info *sinfo, - u64 alloc_bytes) +static int should_alloc_chunk(struct btrfs_root *root, + struct btrfs_space_info *sinfo, u64 alloc_bytes) { u64 num_bytes = sinfo->total_bytes - sinfo->bytes_readonly; + u64 thresh; if (sinfo->bytes_used + sinfo->bytes_reserved + alloc_bytes + 256 * 1024 * 1024 < num_bytes) @@ -3010,6 +3240,12 @@ static int should_alloc_chunk(struct btrfs_space_info *sinfo, alloc_bytes < div_factor(num_bytes, 8)) return 0; + thresh = btrfs_super_total_bytes(&root->fs_info->super_copy); + thresh = max_t(u64, 256 * 1024 * 1024, div_factor_fine(thresh, 5)); + + if (num_bytes > thresh && sinfo->bytes_used < div_factor(num_bytes, 3)) + return 0; + return 1; } @@ -3041,13 +3277,21 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, goto out; } - if (!force && !should_alloc_chunk(space_info, alloc_bytes)) { + if (!force && !should_alloc_chunk(extent_root, space_info, + alloc_bytes)) { spin_unlock(&space_info->lock); goto out; } spin_unlock(&space_info->lock); /* + * If we have mixed data/metadata chunks we want to make sure we keep + * allocating mixed chunks instead of individual chunks. + */ + if (btrfs_mixed_space_info(space_info)) + flags |= (BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA); + + /* * if we're doing a data chunk, go ahead and make sure that * we keep a reasonable number of metadata chunks allocated in the * FS as well. @@ -3072,55 +3316,25 @@ out: return ret; } -static int maybe_allocate_chunk(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_space_info *sinfo, u64 num_bytes) -{ - int ret; - int end_trans = 0; - - if (sinfo->full) - return 0; - - spin_lock(&sinfo->lock); - ret = should_alloc_chunk(sinfo, num_bytes + 2 * 1024 * 1024); - spin_unlock(&sinfo->lock); - if (!ret) - return 0; - - if (!trans) { - trans = btrfs_join_transaction(root, 1); - BUG_ON(IS_ERR(trans)); - end_trans = 1; - } - - ret = do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes + 2 * 1024 * 1024, - get_alloc_profile(root, sinfo->flags), 0); - - if (end_trans) - btrfs_end_transaction(trans, root); - - return ret == 1 ? 1 : 0; -} - /* * shrink metadata reservation for delalloc */ static int shrink_delalloc(struct btrfs_trans_handle *trans, - struct btrfs_root *root, u64 to_reclaim) + struct btrfs_root *root, u64 to_reclaim, int sync) { struct btrfs_block_rsv *block_rsv; + struct btrfs_space_info *space_info; u64 reserved; u64 max_reclaim; u64 reclaimed = 0; int pause = 1; - int ret; + int nr_pages = (2 * 1024 * 1024) >> PAGE_CACHE_SHIFT; block_rsv = &root->fs_info->delalloc_block_rsv; - spin_lock(&block_rsv->lock); - reserved = block_rsv->reserved; - spin_unlock(&block_rsv->lock); + space_info = block_rsv->space_info; + + smp_mb(); + reserved = space_info->bytes_reserved; if (reserved == 0) return 0; @@ -3128,104 +3342,169 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans, max_reclaim = min(reserved, to_reclaim); while (1) { - ret = btrfs_start_one_delalloc_inode(root, trans ? 1 : 0); - if (!ret) { - __set_current_state(TASK_INTERRUPTIBLE); - schedule_timeout(pause); - pause <<= 1; - if (pause > HZ / 10) - pause = HZ / 10; - } else { - pause = 1; - } + /* have the flusher threads jump in and do some IO */ + smp_mb(); + nr_pages = min_t(unsigned long, nr_pages, + root->fs_info->delalloc_bytes >> PAGE_CACHE_SHIFT); + writeback_inodes_sb_nr_if_idle(root->fs_info->sb, nr_pages); - spin_lock(&block_rsv->lock); - if (reserved > block_rsv->reserved) - reclaimed = reserved - block_rsv->reserved; - reserved = block_rsv->reserved; - spin_unlock(&block_rsv->lock); + spin_lock(&space_info->lock); + if (reserved > space_info->bytes_reserved) + reclaimed += reserved - space_info->bytes_reserved; + reserved = space_info->bytes_reserved; + spin_unlock(&space_info->lock); if (reserved == 0 || reclaimed >= max_reclaim) break; if (trans && trans->transaction->blocked) return -EAGAIN; + + __set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(pause); + pause <<= 1; + if (pause > HZ / 10) + pause = HZ / 10; + } return reclaimed >= to_reclaim; } -static int should_retry_reserve(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - struct btrfs_block_rsv *block_rsv, - u64 num_bytes, int *retries) +/* + * Retries tells us how many times we've called reserve_metadata_bytes. The + * idea is if this is the first call (retries == 0) then we will add to our + * reserved count if we can't make the allocation in order to hold our place + * while we go and try and free up space. That way for retries > 1 we don't try + * and add space, we just check to see if the amount of unused space is >= the + * total space, meaning that our reservation is valid. + * + * However if we don't intend to retry this reservation, pass -1 as retries so + * that it short circuits this logic. + */ +static int reserve_metadata_bytes(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + struct btrfs_block_rsv *block_rsv, + u64 orig_bytes, int flush) { struct btrfs_space_info *space_info = block_rsv->space_info; - int ret; + u64 unused; + u64 num_bytes = orig_bytes; + int retries = 0; + int ret = 0; + bool reserved = false; + bool committed = false; - if ((*retries) > 2) - return -ENOSPC; +again: + ret = -ENOSPC; + if (reserved) + num_bytes = 0; - ret = maybe_allocate_chunk(trans, root, space_info, num_bytes); - if (ret) - return 1; + spin_lock(&space_info->lock); + unused = space_info->bytes_used + space_info->bytes_reserved + + space_info->bytes_pinned + space_info->bytes_readonly + + space_info->bytes_may_use; - if (trans && trans->transaction->in_commit) - return -ENOSPC; + /* + * The idea here is that we've not already over-reserved the block group + * then we can go ahead and save our reservation first and then start + * flushing if we need to. Otherwise if we've already overcommitted + * lets start flushing stuff first and then come back and try to make + * our reservation. + */ + if (unused <= space_info->total_bytes) { + unused -= space_info->total_bytes; + if (unused >= num_bytes) { + if (!reserved) + space_info->bytes_reserved += orig_bytes; + ret = 0; + } else { + /* + * Ok set num_bytes to orig_bytes since we aren't + * overocmmitted, this way we only try and reclaim what + * we need. + */ + num_bytes = orig_bytes; + } + } else { + /* + * Ok we're over committed, set num_bytes to the overcommitted + * amount plus the amount of bytes that we need for this + * reservation. + */ + num_bytes = unused - space_info->total_bytes + + (orig_bytes * (retries + 1)); + } - ret = shrink_delalloc(trans, root, num_bytes); - if (ret) - return ret; + /* + * Couldn't make our reservation, save our place so while we're trying + * to reclaim space we can actually use it instead of somebody else + * stealing it from us. + */ + if (ret && !reserved) { + space_info->bytes_reserved += orig_bytes; + reserved = true; + } - spin_lock(&space_info->lock); - if (space_info->bytes_pinned < num_bytes) - ret = 1; spin_unlock(&space_info->lock); - if (ret) - return -ENOSPC; - - (*retries)++; - if (trans) - return -EAGAIN; + if (!ret) + return 0; - trans = btrfs_join_transaction(root, 1); - BUG_ON(IS_ERR(trans)); - ret = btrfs_commit_transaction(trans, root); - BUG_ON(ret); + if (!flush) + goto out; - return 1; -} + /* + * We do synchronous shrinking since we don't actually unreserve + * metadata until after the IO is completed. + */ + ret = shrink_delalloc(trans, root, num_bytes, 1); + if (ret > 0) + return 0; + else if (ret < 0) + goto out; -static int reserve_metadata_bytes(struct btrfs_block_rsv *block_rsv, - u64 num_bytes) -{ - struct btrfs_space_info *space_info = block_rsv->space_info; - u64 unused; - int ret = -ENOSPC; + /* + * So if we were overcommitted it's possible that somebody else flushed + * out enough space and we simply didn't have enough space to reclaim, + * so go back around and try again. + */ + if (retries < 2) { + retries++; + goto again; + } spin_lock(&space_info->lock); - unused = space_info->bytes_used + space_info->bytes_reserved + - space_info->bytes_pinned + space_info->bytes_readonly; + /* + * Not enough space to be reclaimed, don't bother committing the + * transaction. + */ + if (space_info->bytes_pinned < orig_bytes) + ret = -ENOSPC; + spin_unlock(&space_info->lock); + if (ret) + goto out; - if (unused < space_info->total_bytes) - unused = space_info->total_bytes - unused; - else - unused = 0; + ret = -EAGAIN; + if (trans || committed) + goto out; - if (unused >= num_bytes) { - if (block_rsv->priority >= 10) { - space_info->bytes_reserved += num_bytes; - ret = 0; - } else { - if ((unused + block_rsv->reserved) * - block_rsv->priority >= - (num_bytes + block_rsv->reserved) * 10) { - space_info->bytes_reserved += num_bytes; - ret = 0; - } - } + ret = -ENOSPC; + trans = btrfs_join_transaction(root, 1); + if (IS_ERR(trans)) + goto out; + ret = btrfs_commit_transaction(trans, root); + if (!ret) { + trans = NULL; + committed = true; + goto again; + } + +out: + if (reserved) { + spin_lock(&space_info->lock); + space_info->bytes_reserved -= orig_bytes; + spin_unlock(&space_info->lock); } - spin_unlock(&space_info->lock); return ret; } @@ -3327,18 +3606,14 @@ struct btrfs_block_rsv *btrfs_alloc_block_rsv(struct btrfs_root *root) { struct btrfs_block_rsv *block_rsv; struct btrfs_fs_info *fs_info = root->fs_info; - u64 alloc_target; block_rsv = kmalloc(sizeof(*block_rsv), GFP_NOFS); if (!block_rsv) return NULL; btrfs_init_block_rsv(block_rsv); - - alloc_target = btrfs_get_alloc_profile(root, 0); block_rsv->space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); - return block_rsv; } @@ -3369,23 +3644,19 @@ void btrfs_add_durable_block_rsv(struct btrfs_fs_info *fs_info, int btrfs_block_rsv_add(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_block_rsv *block_rsv, - u64 num_bytes, int *retries) + u64 num_bytes) { int ret; if (num_bytes == 0) return 0; -again: - ret = reserve_metadata_bytes(block_rsv, num_bytes); + + ret = reserve_metadata_bytes(trans, root, block_rsv, num_bytes, 1); if (!ret) { block_rsv_add_bytes(block_rsv, num_bytes, 1); return 0; } - ret = should_retry_reserve(trans, root, block_rsv, num_bytes, retries); - if (ret > 0) - goto again; - return ret; } @@ -3420,7 +3691,8 @@ int btrfs_block_rsv_check(struct btrfs_trans_handle *trans, return 0; if (block_rsv->refill_used) { - ret = reserve_metadata_bytes(block_rsv, num_bytes); + ret = reserve_metadata_bytes(trans, root, block_rsv, + num_bytes, 0); if (!ret) { block_rsv_add_bytes(block_rsv, num_bytes, 0); return 0; @@ -3499,6 +3771,8 @@ static u64 calc_global_metadata_size(struct btrfs_fs_info *fs_info) sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); spin_lock(&sinfo->lock); + if (sinfo->flags & BTRFS_BLOCK_GROUP_DATA) + data_used = 0; meta_used = sinfo->bytes_used; spin_unlock(&sinfo->lock); @@ -3526,7 +3800,8 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info) block_rsv->size = num_bytes; num_bytes = sinfo->bytes_used + sinfo->bytes_pinned + - sinfo->bytes_reserved + sinfo->bytes_readonly; + sinfo->bytes_reserved + sinfo->bytes_readonly + + sinfo->bytes_may_use; if (sinfo->total_bytes > num_bytes) { num_bytes = sinfo->total_bytes - num_bytes; @@ -3597,7 +3872,7 @@ static u64 calc_trans_metadata_size(struct btrfs_root *root, int num_items) int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans, struct btrfs_root *root, - int num_items, int *retries) + int num_items) { u64 num_bytes; int ret; @@ -3607,7 +3882,7 @@ int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans, num_bytes = calc_trans_metadata_size(root, num_items); ret = btrfs_block_rsv_add(trans, root, &root->fs_info->trans_block_rsv, - num_bytes, retries); + num_bytes); if (!ret) { trans->bytes_reserved += num_bytes; trans->block_rsv = &root->fs_info->trans_block_rsv; @@ -3681,14 +3956,13 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) struct btrfs_block_rsv *block_rsv = &root->fs_info->delalloc_block_rsv; u64 to_reserve; int nr_extents; - int retries = 0; int ret; if (btrfs_transaction_in_commit(root->fs_info)) schedule_timeout(1); num_bytes = ALIGN(num_bytes, root->sectorsize); -again: + spin_lock(&BTRFS_I(inode)->accounting_lock); nr_extents = atomic_read(&BTRFS_I(inode)->outstanding_extents) + 1; if (nr_extents > BTRFS_I(inode)->reserved_extents) { @@ -3698,18 +3972,14 @@ again: nr_extents = 0; to_reserve = 0; } + spin_unlock(&BTRFS_I(inode)->accounting_lock); to_reserve += calc_csum_metadata_size(inode, num_bytes); - ret = reserve_metadata_bytes(block_rsv, to_reserve); - if (ret) { - spin_unlock(&BTRFS_I(inode)->accounting_lock); - ret = should_retry_reserve(NULL, root, block_rsv, to_reserve, - &retries); - if (ret > 0) - goto again; + ret = reserve_metadata_bytes(NULL, root, block_rsv, to_reserve, 1); + if (ret) return ret; - } + spin_lock(&BTRFS_I(inode)->accounting_lock); BTRFS_I(inode)->reserved_extents += nr_extents; atomic_inc(&BTRFS_I(inode)->outstanding_extents); spin_unlock(&BTRFS_I(inode)->accounting_lock); @@ -3717,7 +3987,7 @@ again: block_rsv_add_bytes(block_rsv, to_reserve, 1); if (block_rsv->size > 512 * 1024 * 1024) - shrink_delalloc(NULL, root, to_reserve); + shrink_delalloc(NULL, root, to_reserve, 0); return 0; } @@ -3776,12 +4046,12 @@ static int update_block_group(struct btrfs_trans_handle *trans, struct btrfs_root *root, u64 bytenr, u64 num_bytes, int alloc) { - struct btrfs_block_group_cache *cache; + struct btrfs_block_group_cache *cache = NULL; struct btrfs_fs_info *info = root->fs_info; - int factor; u64 total = num_bytes; u64 old_val; u64 byte_in_group; + int factor; /* block accounting for super block */ spin_lock(&info->delalloc_lock); @@ -3803,11 +4073,25 @@ static int update_block_group(struct btrfs_trans_handle *trans, factor = 2; else factor = 1; + /* + * If this block group has free space cache written out, we + * need to make sure to load it if we are removing space. This + * is because we need the unpinning stage to actually add the + * space back to the block group, otherwise we will leak space. + */ + if (!alloc && cache->cached == BTRFS_CACHE_NO) + cache_block_group(cache, trans, 1); + byte_in_group = bytenr - cache->key.objectid; WARN_ON(byte_in_group > cache->key.offset); spin_lock(&cache->space_info->lock); spin_lock(&cache->lock); + + if (btrfs_super_cache_generation(&info->super_copy) != 0 && + cache->disk_cache_state < BTRFS_DC_CLEAR) + cache->disk_cache_state = BTRFS_DC_CLEAR; + cache->dirty = 1; old_val = btrfs_block_group_used(&cache->item); num_bytes = min(total, cache->key.offset - byte_in_group); @@ -4554,6 +4838,7 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, bool found_uncached_bg = false; bool failed_cluster_refill = false; bool failed_alloc = false; + bool use_cluster = true; u64 ideal_cache_percent = 0; u64 ideal_cache_offset = 0; @@ -4568,16 +4853,24 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans, return -ENOSPC; } + /* + * If the space info is for both data and metadata it means we have a + * small filesystem and we can't use the clustering stuff. + */ + if (btrfs_mixed_space_info(space_info)) + use_cluster = false; + if (orig_root->ref_cows || empty_size) allowed_chunk_alloc = 1; - if (data & BTRFS_BLOCK_GROUP_METADATA) { + if (data & BTRFS_BLOCK_GROUP_METADATA && use_cluster) { last_ptr = &root->fs_info->meta_alloc_cluster; if (!btrfs_test_opt(root, SSD)) empty_cluster = 64 * 1024; } - if ((data & BTRFS_BLOCK_GROUP_DATA) && btrfs_test_opt(root, SSD)) { + if ((data & BTRFS_BLOCK_GROUP_DATA) && use_cluster && + btrfs_test_opt(root, SSD)) { last_ptr = &root->fs_info->data_alloc_cluster; } @@ -4641,6 +4934,10 @@ have_block_group: if (unlikely(block_group->cached == BTRFS_CACHE_NO)) { u64 free_percent; + ret = cache_block_group(block_group, trans, 1); + if (block_group->cached == BTRFS_CACHE_FINISHED) + goto have_block_group; + free_percent = btrfs_block_group_used(&block_group->item); free_percent *= 100; free_percent = div64_u64(free_percent, @@ -4661,7 +4958,7 @@ have_block_group: if (loop > LOOP_CACHING_NOWAIT || (loop > LOOP_FIND_IDEAL && atomic_read(&space_info->caching_threads) < 2)) { - ret = cache_block_group(block_group); + ret = cache_block_group(block_group, trans, 0); BUG_ON(ret); } found_uncached_bg = true; @@ -5218,7 +5515,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans, u64 num_bytes = ins->offset; block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid); - cache_block_group(block_group); + cache_block_group(block_group, trans, 0); caching_ctl = get_caching_control(block_group); if (!caching_ctl) { @@ -5308,7 +5605,8 @@ use_block_rsv(struct btrfs_trans_handle *trans, block_rsv = get_block_rsv(trans, root); if (block_rsv->size == 0) { - ret = reserve_metadata_bytes(block_rsv, blocksize); + ret = reserve_metadata_bytes(trans, root, block_rsv, + blocksize, 0); if (ret) return ERR_PTR(ret); return block_rsv; @@ -5318,11 +5616,6 @@ use_block_rsv(struct btrfs_trans_handle *trans, if (!ret) return block_rsv; - WARN_ON(1); - printk(KERN_INFO"block_rsv size %llu reserved %llu freed %llu %llu\n", - block_rsv->size, block_rsv->reserved, - block_rsv->freed[0], block_rsv->freed[1]); - return ERR_PTR(-ENOSPC); } @@ -5421,7 +5714,6 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans, u64 generation; u64 refs; u64 flags; - u64 last = 0; u32 nritems; u32 blocksize; struct btrfs_key key; @@ -5489,7 +5781,6 @@ reada: generation); if (ret) break; - last = bytenr + blocksize; nread++; } wc->reada_slot = slot; @@ -7813,6 +8104,40 @@ out: return ret; } +void btrfs_put_block_group_cache(struct btrfs_fs_info *info) +{ + struct btrfs_block_group_cache *block_group; + u64 last = 0; + + while (1) { + struct inode *inode; + + block_group = btrfs_lookup_first_block_group(info, last); + while (block_group) { + spin_lock(&block_group->lock); + if (block_group->iref) + break; + spin_unlock(&block_group->lock); + block_group = next_block_group(info->tree_root, + block_group); + } + if (!block_group) { + if (last == 0) + break; + last = 0; + continue; + } + + inode = block_group->inode; + block_group->iref = 0; + block_group->inode = NULL; + spin_unlock(&block_group->lock); + iput(inode); + last = block_group->key.objectid + block_group->key.offset; + btrfs_put_block_group(block_group); + } +} + int btrfs_free_block_groups(struct btrfs_fs_info *info) { struct btrfs_block_group_cache *block_group; @@ -7896,6 +8221,8 @@ int btrfs_read_block_groups(struct btrfs_root *root) struct btrfs_key key; struct btrfs_key found_key; struct extent_buffer *leaf; + int need_clear = 0; + u64 cache_gen; root = info->extent_root; key.objectid = 0; @@ -7905,6 +8232,15 @@ int btrfs_read_block_groups(struct btrfs_root *root) if (!path) return -ENOMEM; + cache_gen = btrfs_super_cache_generation(&root->fs_info->super_copy); + if (cache_gen != 0 && + btrfs_super_generation(&root->fs_info->super_copy) != cache_gen) + need_clear = 1; + if (btrfs_test_opt(root, CLEAR_CACHE)) + need_clear = 1; + if (!btrfs_test_opt(root, SPACE_CACHE) && cache_gen) + printk(KERN_INFO "btrfs: disk space caching is enabled\n"); + while (1) { ret = find_first_block_group(root, path, &key); if (ret > 0) @@ -7927,6 +8263,9 @@ int btrfs_read_block_groups(struct btrfs_root *root) INIT_LIST_HEAD(&cache->list); INIT_LIST_HEAD(&cache->cluster_list); + if (need_clear) + cache->disk_cache_state = BTRFS_DC_CLEAR; + /* * we only want to have 32k of ram per block group for keeping * track of free space, and if we pass 1/2 of that we want to @@ -8031,6 +8370,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, cache->key.offset = size; cache->key.type = BTRFS_BLOCK_GROUP_ITEM_KEY; cache->sectorsize = root->sectorsize; + cache->fs_info = root->fs_info; /* * we only want to have 32k of ram per block group for keeping track @@ -8087,8 +8427,11 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct btrfs_block_group_cache *block_group; struct btrfs_free_cluster *cluster; + struct btrfs_root *tree_root = root->fs_info->tree_root; struct btrfs_key key; + struct inode *inode; int ret; + int factor; root = root->fs_info->extent_root; @@ -8097,6 +8440,12 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, BUG_ON(!block_group->ro); memcpy(&key, &block_group->key, sizeof(key)); + if (block_group->flags & (BTRFS_BLOCK_GROUP_DUP | + BTRFS_BLOCK_GROUP_RAID1 | + BTRFS_BLOCK_GROUP_RAID10)) + factor = 2; + else + factor = 1; /* make sure this block group isn't part of an allocation cluster */ cluster = &root->fs_info->data_alloc_cluster; @@ -8116,6 +8465,40 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, path = btrfs_alloc_path(); BUG_ON(!path); + inode = lookup_free_space_inode(root, block_group, path); + if (!IS_ERR(inode)) { + btrfs_orphan_add(trans, inode); + clear_nlink(inode); + /* One for the block groups ref */ + spin_lock(&block_group->lock); + if (block_group->iref) { + block_group->iref = 0; + block_group->inode = NULL; + spin_unlock(&block_group->lock); + iput(inode); + } else { + spin_unlock(&block_group->lock); + } + /* One for our lookup ref */ + iput(inode); + } + + key.objectid = BTRFS_FREE_SPACE_OBJECTID; + key.offset = block_group->key.objectid; + key.type = 0; + + ret = btrfs_search_slot(trans, tree_root, &key, path, -1, 1); + if (ret < 0) + goto out; + if (ret > 0) + btrfs_release_path(tree_root, path); + if (ret == 0) { + ret = btrfs_del_item(trans, tree_root, path); + if (ret) + goto out; + btrfs_release_path(tree_root, path); + } + spin_lock(&root->fs_info->block_group_cache_lock); rb_erase(&block_group->cache_node, &root->fs_info->block_group_cache_tree); @@ -8137,8 +8520,11 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, spin_lock(&block_group->space_info->lock); block_group->space_info->total_bytes -= block_group->key.offset; block_group->space_info->bytes_readonly -= block_group->key.offset; + block_group->space_info->disk_total -= block_group->key.offset * factor; spin_unlock(&block_group->space_info->lock); + memcpy(&key, &block_group->key, sizeof(key)); + btrfs_clear_space_info_full(root->fs_info); btrfs_put_block_group(block_group); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d74e6af..eac10e3 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -104,7 +104,7 @@ void extent_io_tree_init(struct extent_io_tree *tree, struct address_space *mapping, gfp_t mask) { tree->state = RB_ROOT; - tree->buffer = RB_ROOT; + INIT_RADIX_TREE(&tree->buffer, GFP_ATOMIC); tree->ops = NULL; tree->dirty_bytes = 0; spin_lock_init(&tree->lock); @@ -235,50 +235,6 @@ static inline struct rb_node *tree_search(struct extent_io_tree *tree, return ret; } -static struct extent_buffer *buffer_tree_insert(struct extent_io_tree *tree, - u64 offset, struct rb_node *node) -{ - struct rb_root *root = &tree->buffer; - struct rb_node **p = &root->rb_node; - struct rb_node *parent = NULL; - struct extent_buffer *eb; - - while (*p) { - parent = *p; - eb = rb_entry(parent, struct extent_buffer, rb_node); - - if (offset < eb->start) - p = &(*p)->rb_left; - else if (offset > eb->start) - p = &(*p)->rb_right; - else - return eb; - } - - rb_link_node(node, parent, p); - rb_insert_color(node, root); - return NULL; -} - -static struct extent_buffer *buffer_search(struct extent_io_tree *tree, - u64 offset) -{ - struct rb_root *root = &tree->buffer; - struct rb_node *n = root->rb_node; - struct extent_buffer *eb; - - while (n) { - eb = rb_entry(n, struct extent_buffer, rb_node); - if (offset < eb->start) - n = n->rb_left; - else if (offset > eb->start) - n = n->rb_right; - else - return eb; - } - return NULL; -} - static void merge_cb(struct extent_io_tree *tree, struct extent_state *new, struct extent_state *other) { @@ -1901,10 +1857,8 @@ static int submit_one_bio(int rw, struct bio *bio, int mirror_num, struct page *page = bvec->bv_page; struct extent_io_tree *tree = bio->bi_private; u64 start; - u64 end; start = ((u64)page->index << PAGE_CACHE_SHIFT) + bvec->bv_offset; - end = start + bvec->bv_len - 1; bio->bi_private = NULL; @@ -2204,7 +2158,6 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, u64 last_byte = i_size_read(inode); u64 block_start; u64 iosize; - u64 unlock_start; sector_t sector; struct extent_state *cached_state = NULL; struct extent_map *em; @@ -2329,7 +2282,6 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, if (tree->ops && tree->ops->writepage_end_io_hook) tree->ops->writepage_end_io_hook(page, start, page_end, NULL, 1); - unlock_start = page_end + 1; goto done; } @@ -2340,7 +2292,6 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, if (tree->ops && tree->ops->writepage_end_io_hook) tree->ops->writepage_end_io_hook(page, cur, page_end, NULL, 1); - unlock_start = page_end + 1; break; } em = epd->get_extent(inode, page, pg_offset, cur, @@ -2387,7 +2338,6 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, cur += iosize; pg_offset += iosize; - unlock_start = cur; continue; } /* leave this out until we have a page_mkwrite call */ @@ -2473,7 +2423,6 @@ static int extent_write_cache_pages(struct extent_io_tree *tree, pgoff_t index; pgoff_t end; /* Inclusive */ int scanned = 0; - int range_whole = 0; pagevec_init(&pvec, 0); if (wbc->range_cyclic) { @@ -2482,8 +2431,6 @@ static int extent_write_cache_pages(struct extent_io_tree *tree, } else { index = wbc->range_start >> PAGE_CACHE_SHIFT; end = wbc->range_end >> PAGE_CACHE_SHIFT; - if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) - range_whole = 1; scanned = 1; } retry: @@ -2823,6 +2770,8 @@ int extent_prepare_write(struct extent_io_tree *tree, NULL, 1, end_bio_extent_preparewrite, 0, 0, 0); + if (ret && !err) + err = ret; iocount++; block_start = block_start + iosize; } else { @@ -3104,6 +3053,39 @@ static void __free_extent_buffer(struct extent_buffer *eb) kmem_cache_free(extent_buffer_cache, eb); } +/* + * Helper for releasing extent buffer page. + */ +static void btrfs_release_extent_buffer_page(struct extent_buffer *eb, + unsigned long start_idx) +{ + unsigned long index; + struct page *page; + + if (!eb->first_page) + return; + + index = num_extent_pages(eb->start, eb->len); + if (start_idx >= index) + return; + + do { + index--; + page = extent_buffer_page(eb, index); + if (page) + page_cache_release(page); + } while (index != start_idx); +} + +/* + * Helper for releasing the extent buffer. + */ +static inline void btrfs_release_extent_buffer(struct extent_buffer *eb) +{ + btrfs_release_extent_buffer_page(eb, 0); + __free_extent_buffer(eb); +} + struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, u64 start, unsigned long len, struct page *page0, @@ -3117,16 +3099,16 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, struct page *p; struct address_space *mapping = tree->mapping; int uptodate = 1; + int ret; - spin_lock(&tree->buffer_lock); - eb = buffer_search(tree, start); - if (eb) { - atomic_inc(&eb->refs); - spin_unlock(&tree->buffer_lock); + rcu_read_lock(); + eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); + if (eb && atomic_inc_not_zero(&eb->refs)) { + rcu_read_unlock(); mark_page_accessed(eb->first_page); return eb; } - spin_unlock(&tree->buffer_lock); + rcu_read_unlock(); eb = __alloc_extent_buffer(tree, start, len, mask); if (!eb) @@ -3165,26 +3147,31 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, if (uptodate) set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); + ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM); + if (ret) + goto free_eb; + spin_lock(&tree->buffer_lock); - exists = buffer_tree_insert(tree, start, &eb->rb_node); - if (exists) { + ret = radix_tree_insert(&tree->buffer, start >> PAGE_CACHE_SHIFT, eb); + if (ret == -EEXIST) { + exists = radix_tree_lookup(&tree->buffer, + start >> PAGE_CACHE_SHIFT); /* add one reference for the caller */ atomic_inc(&exists->refs); spin_unlock(&tree->buffer_lock); + radix_tree_preload_end(); goto free_eb; } /* add one reference for the tree */ atomic_inc(&eb->refs); spin_unlock(&tree->buffer_lock); + radix_tree_preload_end(); return eb; free_eb: if (!atomic_dec_and_test(&eb->refs)) return exists; - for (index = 1; index < i; index++) - page_cache_release(extent_buffer_page(eb, index)); - page_cache_release(extent_buffer_page(eb, 0)); - __free_extent_buffer(eb); + btrfs_release_extent_buffer(eb); return exists; } @@ -3194,16 +3181,16 @@ struct extent_buffer *find_extent_buffer(struct extent_io_tree *tree, { struct extent_buffer *eb; - spin_lock(&tree->buffer_lock); - eb = buffer_search(tree, start); - if (eb) - atomic_inc(&eb->refs); - spin_unlock(&tree->buffer_lock); - - if (eb) + rcu_read_lock(); + eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); + if (eb && atomic_inc_not_zero(&eb->refs)) { + rcu_read_unlock(); mark_page_accessed(eb->first_page); + return eb; + } + rcu_read_unlock(); - return eb; + return NULL; } void free_extent_buffer(struct extent_buffer *eb) @@ -3833,34 +3820,45 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset, } } +static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head) +{ + struct extent_buffer *eb = + container_of(head, struct extent_buffer, rcu_head); + + btrfs_release_extent_buffer(eb); +} + int try_release_extent_buffer(struct extent_io_tree *tree, struct page *page) { u64 start = page_offset(page); struct extent_buffer *eb; int ret = 1; - unsigned long i; - unsigned long num_pages; spin_lock(&tree->buffer_lock); - eb = buffer_search(tree, start); + eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); if (!eb) goto out; - if (atomic_read(&eb->refs) > 1) { + if (test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) { ret = 0; goto out; } - if (test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) { + + /* + * set @eb->refs to 0 if it is already 1, and then release the @eb. + * Or go back. + */ + if (atomic_cmpxchg(&eb->refs, 1, 0) != 1) { ret = 0; goto out; } - /* at this point we can safely release the extent buffer */ - num_pages = num_extent_pages(eb->start, eb->len); - for (i = 0; i < num_pages; i++) - page_cache_release(extent_buffer_page(eb, i)); - rb_erase(&eb->rb_node, &tree->buffer); - __free_extent_buffer(eb); + + radix_tree_delete(&tree->buffer, start >> PAGE_CACHE_SHIFT); out: spin_unlock(&tree->buffer_lock); + + /* at this point we can safely release the extent buffer */ + if (atomic_read(&eb->refs) == 0) + call_rcu(&eb->rcu_head, btrfs_release_extent_buffer_rcu); return ret; } diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 5691c7b..1c6d4f3 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -85,7 +85,7 @@ struct extent_io_ops { struct extent_io_tree { struct rb_root state; - struct rb_root buffer; + struct radix_tree_root buffer; struct address_space *mapping; u64 dirty_bytes; spinlock_t lock; @@ -123,7 +123,7 @@ struct extent_buffer { unsigned long bflags; atomic_t refs; struct list_head leak_list; - struct rb_node rb_node; + struct rcu_head rcu_head; /* the spinlock is used to protect most operations */ spinlock_t lock; diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index 454ca52..23cb8da 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -335,7 +335,7 @@ struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree, goto out; } if (IS_ERR(rb_node)) { - em = ERR_PTR(PTR_ERR(rb_node)); + em = ERR_CAST(rb_node); goto out; } em = rb_entry(rb_node, struct extent_map, rb_node); @@ -384,7 +384,7 @@ struct extent_map *search_extent_mapping(struct extent_map_tree *tree, goto out; } if (IS_ERR(rb_node)) { - em = ERR_PTR(PTR_ERR(rb_node)); + em = ERR_CAST(rb_node); goto out; } em = rb_entry(rb_node, struct extent_map, rb_node); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index f488fac..22ee0dc 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -23,10 +23,761 @@ #include "ctree.h" #include "free-space-cache.h" #include "transaction.h" +#include "disk-io.h" #define BITS_PER_BITMAP (PAGE_CACHE_SIZE * 8) #define MAX_CACHE_BYTES_PER_GIG (32 * 1024) +static void recalculate_thresholds(struct btrfs_block_group_cache + *block_group); +static int link_free_space(struct btrfs_block_group_cache *block_group, + struct btrfs_free_space *info); + +struct inode *lookup_free_space_inode(struct btrfs_root *root, + struct btrfs_block_group_cache + *block_group, struct btrfs_path *path) +{ + struct btrfs_key key; + struct btrfs_key location; + struct btrfs_disk_key disk_key; + struct btrfs_free_space_header *header; + struct extent_buffer *leaf; + struct inode *inode = NULL; + int ret; + + spin_lock(&block_group->lock); + if (block_group->inode) + inode = igrab(block_group->inode); + spin_unlock(&block_group->lock); + if (inode) + return inode; + + key.objectid = BTRFS_FREE_SPACE_OBJECTID; + key.offset = block_group->key.objectid; + key.type = 0; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + return ERR_PTR(ret); + if (ret > 0) { + btrfs_release_path(root, path); + return ERR_PTR(-ENOENT); + } + + leaf = path->nodes[0]; + header = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_free_space_header); + btrfs_free_space_key(leaf, header, &disk_key); + btrfs_disk_key_to_cpu(&location, &disk_key); + btrfs_release_path(root, path); + + inode = btrfs_iget(root->fs_info->sb, &location, root, NULL); + if (!inode) + return ERR_PTR(-ENOENT); + if (IS_ERR(inode)) + return inode; + if (is_bad_inode(inode)) { + iput(inode); + return ERR_PTR(-ENOENT); + } + + spin_lock(&block_group->lock); + if (!root->fs_info->closing) { + block_group->inode = igrab(inode); + block_group->iref = 1; + } + spin_unlock(&block_group->lock); + + return inode; +} + +int create_free_space_inode(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_block_group_cache *block_group, + struct btrfs_path *path) +{ + struct btrfs_key key; + struct btrfs_disk_key disk_key; + struct btrfs_free_space_header *header; + struct btrfs_inode_item *inode_item; + struct extent_buffer *leaf; + u64 objectid; + int ret; + + ret = btrfs_find_free_objectid(trans, root, 0, &objectid); + if (ret < 0) + return ret; + + ret = btrfs_insert_empty_inode(trans, root, path, objectid); + if (ret) + return ret; + + leaf = path->nodes[0]; + inode_item = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_inode_item); + btrfs_item_key(leaf, &disk_key, path->slots[0]); + memset_extent_buffer(leaf, 0, (unsigned long)inode_item, + sizeof(*inode_item)); + btrfs_set_inode_generation(leaf, inode_item, trans->transid); + btrfs_set_inode_size(leaf, inode_item, 0); + btrfs_set_inode_nbytes(leaf, inode_item, 0); + btrfs_set_inode_uid(leaf, inode_item, 0); + btrfs_set_inode_gid(leaf, inode_item, 0); + btrfs_set_inode_mode(leaf, inode_item, S_IFREG | 0600); + btrfs_set_inode_flags(leaf, inode_item, BTRFS_INODE_NOCOMPRESS | + BTRFS_INODE_PREALLOC | BTRFS_INODE_NODATASUM); + btrfs_set_inode_nlink(leaf, inode_item, 1); + btrfs_set_inode_transid(leaf, inode_item, trans->transid); + btrfs_set_inode_block_group(leaf, inode_item, + block_group->key.objectid); + btrfs_mark_buffer_dirty(leaf); + btrfs_release_path(root, path); + + key.objectid = BTRFS_FREE_SPACE_OBJECTID; + key.offset = block_group->key.objectid; + key.type = 0; + + ret = btrfs_insert_empty_item(trans, root, path, &key, + sizeof(struct btrfs_free_space_header)); + if (ret < 0) { + btrfs_release_path(root, path); + return ret; + } + leaf = path->nodes[0]; + header = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_free_space_header); + memset_extent_buffer(leaf, 0, (unsigned long)header, sizeof(*header)); + btrfs_set_free_space_key(leaf, header, &disk_key); + btrfs_mark_buffer_dirty(leaf); + btrfs_release_path(root, path); + + return 0; +} + +int btrfs_truncate_free_space_cache(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_path *path, + struct inode *inode) +{ + loff_t oldsize; + int ret = 0; + + trans->block_rsv = root->orphan_block_rsv; + ret = btrfs_block_rsv_check(trans, root, + root->orphan_block_rsv, + 0, 5); + if (ret) + return ret; + + oldsize = i_size_read(inode); + btrfs_i_size_write(inode, 0); + truncate_pagecache(inode, oldsize, 0); + + /* + * We don't need an orphan item because truncating the free space cache + * will never be split across transactions. + */ + ret = btrfs_truncate_inode_items(trans, root, inode, + 0, BTRFS_EXTENT_DATA_KEY); + if (ret) { + WARN_ON(1); + return ret; + } + + return btrfs_update_inode(trans, root, inode); +} + +static int readahead_cache(struct inode *inode) +{ + struct file_ra_state *ra; + unsigned long last_index; + + ra = kzalloc(sizeof(*ra), GFP_NOFS); + if (!ra) + return -ENOMEM; + + file_ra_state_init(ra, inode->i_mapping); + last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT; + + page_cache_sync_readahead(inode->i_mapping, ra, NULL, 0, last_index); + + kfree(ra); + + return 0; +} + +int load_free_space_cache(struct btrfs_fs_info *fs_info, + struct btrfs_block_group_cache *block_group) +{ + struct btrfs_root *root = fs_info->tree_root; + struct inode *inode; + struct btrfs_free_space_header *header; + struct extent_buffer *leaf; + struct page *page; + struct btrfs_path *path; + u32 *checksums = NULL, *crc; + char *disk_crcs = NULL; + struct btrfs_key key; + struct list_head bitmaps; + u64 num_entries; + u64 num_bitmaps; + u64 generation; + u32 cur_crc = ~(u32)0; + pgoff_t index = 0; + unsigned long first_page_offset; + int num_checksums; + int ret = 0; + + /* + * If we're unmounting then just return, since this does a search on the + * normal root and not the commit root and we could deadlock. + */ + smp_mb(); + if (fs_info->closing) + return 0; + + /* + * If this block group has been marked to be cleared for one reason or + * another then we can't trust the on disk cache, so just return. + */ + spin_lock(&block_group->lock); + if (block_group->disk_cache_state != BTRFS_DC_WRITTEN) { + spin_unlock(&block_group->lock); + return 0; + } + spin_unlock(&block_group->lock); + + INIT_LIST_HEAD(&bitmaps); + + path = btrfs_alloc_path(); + if (!path) + return 0; + + inode = lookup_free_space_inode(root, block_group, path); + if (IS_ERR(inode)) { + btrfs_free_path(path); + return 0; + } + + /* Nothing in the space cache, goodbye */ + if (!i_size_read(inode)) { + btrfs_free_path(path); + goto out; + } + + key.objectid = BTRFS_FREE_SPACE_OBJECTID; + key.offset = block_group->key.objectid; + key.type = 0; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret) { + btrfs_free_path(path); + goto out; + } + + leaf = path->nodes[0]; + header = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_free_space_header); + num_entries = btrfs_free_space_entries(leaf, header); + num_bitmaps = btrfs_free_space_bitmaps(leaf, header); + generation = btrfs_free_space_generation(leaf, header); + btrfs_free_path(path); + + if (BTRFS_I(inode)->generation != generation) { + printk(KERN_ERR "btrfs: free space inode generation (%llu) did" + " not match free space cache generation (%llu) for " + "block group %llu\n", + (unsigned long long)BTRFS_I(inode)->generation, + (unsigned long long)generation, + (unsigned long long)block_group->key.objectid); + goto out; + } + + if (!num_entries) + goto out; + + /* Setup everything for doing checksumming */ + num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE; + checksums = crc = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS); + if (!checksums) + goto out; + first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64); + disk_crcs = kzalloc(first_page_offset, GFP_NOFS); + if (!disk_crcs) + goto out; + + ret = readahead_cache(inode); + if (ret) { + ret = 0; + goto out; + } + + while (1) { + struct btrfs_free_space_entry *entry; + struct btrfs_free_space *e; + void *addr; + unsigned long offset = 0; + unsigned long start_offset = 0; + int need_loop = 0; + + if (!num_entries && !num_bitmaps) + break; + + if (index == 0) { + start_offset = first_page_offset; + offset = start_offset; + } + + page = grab_cache_page(inode->i_mapping, index); + if (!page) { + ret = 0; + goto free_cache; + } + + if (!PageUptodate(page)) { + btrfs_readpage(NULL, page); + lock_page(page); + if (!PageUptodate(page)) { + unlock_page(page); + page_cache_release(page); + printk(KERN_ERR "btrfs: error reading free " + "space cache: %llu\n", + (unsigned long long) + block_group->key.objectid); + goto free_cache; + } + } + addr = kmap(page); + + if (index == 0) { + u64 *gen; + + memcpy(disk_crcs, addr, first_page_offset); + gen = addr + (sizeof(u32) * num_checksums); + if (*gen != BTRFS_I(inode)->generation) { + printk(KERN_ERR "btrfs: space cache generation" + " (%llu) does not match inode (%llu) " + "for block group %llu\n", + (unsigned long long)*gen, + (unsigned long long) + BTRFS_I(inode)->generation, + (unsigned long long) + block_group->key.objectid); + kunmap(page); + unlock_page(page); + page_cache_release(page); + goto free_cache; + } + crc = (u32 *)disk_crcs; + } + entry = addr + start_offset; + + /* First lets check our crc before we do anything fun */ + cur_crc = ~(u32)0; + cur_crc = btrfs_csum_data(root, addr + start_offset, cur_crc, + PAGE_CACHE_SIZE - start_offset); + btrfs_csum_final(cur_crc, (char *)&cur_crc); + if (cur_crc != *crc) { + printk(KERN_ERR "btrfs: crc mismatch for page %lu in " + "block group %llu\n", index, + (unsigned long long)block_group->key.objectid); + kunmap(page); + unlock_page(page); + page_cache_release(page); + goto free_cache; + } + crc++; + + while (1) { + if (!num_entries) + break; + + need_loop = 1; + e = kzalloc(sizeof(struct btrfs_free_space), GFP_NOFS); + if (!e) { + kunmap(page); + unlock_page(page); + page_cache_release(page); + goto free_cache; + } + + e->offset = le64_to_cpu(entry->offset); + e->bytes = le64_to_cpu(entry->bytes); + if (!e->bytes) { + kunmap(page); + kfree(e); + unlock_page(page); + page_cache_release(page); + goto free_cache; + } + + if (entry->type == BTRFS_FREE_SPACE_EXTENT) { + spin_lock(&block_group->tree_lock); + ret = link_free_space(block_group, e); + spin_unlock(&block_group->tree_lock); + BUG_ON(ret); + } else { + e->bitmap = kzalloc(PAGE_CACHE_SIZE, GFP_NOFS); + if (!e->bitmap) { + kunmap(page); + kfree(e); + unlock_page(page); + page_cache_release(page); + goto free_cache; + } + spin_lock(&block_group->tree_lock); + ret = link_free_space(block_group, e); + block_group->total_bitmaps++; + recalculate_thresholds(block_group); + spin_unlock(&block_group->tree_lock); + list_add_tail(&e->list, &bitmaps); + } + + num_entries--; + offset += sizeof(struct btrfs_free_space_entry); + if (offset + sizeof(struct btrfs_free_space_entry) >= + PAGE_CACHE_SIZE) + break; + entry++; + } + + /* + * We read an entry out of this page, we need to move on to the + * next page. + */ + if (need_loop) { + kunmap(page); + goto next; + } + + /* + * We add the bitmaps at the end of the entries in order that + * the bitmap entries are added to the cache. + */ + e = list_entry(bitmaps.next, struct btrfs_free_space, list); + list_del_init(&e->list); + memcpy(e->bitmap, addr, PAGE_CACHE_SIZE); + kunmap(page); + num_bitmaps--; +next: + unlock_page(page); + page_cache_release(page); + index++; + } + + ret = 1; +out: + kfree(checksums); + kfree(disk_crcs); + iput(inode); + return ret; + +free_cache: + /* This cache is bogus, make sure it gets cleared */ + spin_lock(&block_group->lock); + block_group->disk_cache_state = BTRFS_DC_CLEAR; + spin_unlock(&block_group->lock); + btrfs_remove_free_space_cache(block_group); + goto out; +} + +int btrfs_write_out_cache(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_block_group_cache *block_group, + struct btrfs_path *path) +{ + struct btrfs_free_space_header *header; + struct extent_buffer *leaf; + struct inode *inode; + struct rb_node *node; + struct list_head *pos, *n; + struct page *page; + struct extent_state *cached_state = NULL; + struct list_head bitmap_list; + struct btrfs_key key; + u64 bytes = 0; + u32 *crc, *checksums; + pgoff_t index = 0, last_index = 0; + unsigned long first_page_offset; + int num_checksums; + int entries = 0; + int bitmaps = 0; + int ret = 0; + + root = root->fs_info->tree_root; + + INIT_LIST_HEAD(&bitmap_list); + + spin_lock(&block_group->lock); + if (block_group->disk_cache_state < BTRFS_DC_SETUP) { + spin_unlock(&block_group->lock); + return 0; + } + spin_unlock(&block_group->lock); + + inode = lookup_free_space_inode(root, block_group, path); + if (IS_ERR(inode)) + return 0; + + if (!i_size_read(inode)) { + iput(inode); + return 0; + } + + last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT; + filemap_write_and_wait(inode->i_mapping); + btrfs_wait_ordered_range(inode, inode->i_size & + ~(root->sectorsize - 1), (u64)-1); + + /* We need a checksum per page. */ + num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE; + crc = checksums = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS); + if (!crc) { + iput(inode); + return 0; + } + + /* Since the first page has all of our checksums and our generation we + * need to calculate the offset into the page that we can start writing + * our entries. + */ + first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64); + + node = rb_first(&block_group->free_space_offset); + if (!node) + goto out_free; + + /* + * Lock all pages first so we can lock the extent safely. + * + * NOTE: Because we hold the ref the entire time we're going to write to + * the page find_get_page should never fail, so we don't do a check + * after find_get_page at this point. Just putting this here so people + * know and don't freak out. + */ + while (index <= last_index) { + page = grab_cache_page(inode->i_mapping, index); + if (!page) { + pgoff_t i = 0; + + while (i < index) { + page = find_get_page(inode->i_mapping, i); + unlock_page(page); + page_cache_release(page); + page_cache_release(page); + i++; + } + goto out_free; + } + index++; + } + + index = 0; + lock_extent_bits(&BTRFS_I(inode)->io_tree, 0, i_size_read(inode) - 1, + 0, &cached_state, GFP_NOFS); + + /* Write out the extent entries */ + do { + struct btrfs_free_space_entry *entry; + void *addr; + unsigned long offset = 0; + unsigned long start_offset = 0; + + if (index == 0) { + start_offset = first_page_offset; + offset = start_offset; + } + + page = find_get_page(inode->i_mapping, index); + + addr = kmap(page); + entry = addr + start_offset; + + memset(addr, 0, PAGE_CACHE_SIZE); + while (1) { + struct btrfs_free_space *e; + + e = rb_entry(node, struct btrfs_free_space, offset_index); + entries++; + + entry->offset = cpu_to_le64(e->offset); + entry->bytes = cpu_to_le64(e->bytes); + if (e->bitmap) { + entry->type = BTRFS_FREE_SPACE_BITMAP; + list_add_tail(&e->list, &bitmap_list); + bitmaps++; + } else { + entry->type = BTRFS_FREE_SPACE_EXTENT; + } + node = rb_next(node); + if (!node) + break; + offset += sizeof(struct btrfs_free_space_entry); + if (offset + sizeof(struct btrfs_free_space_entry) >= + PAGE_CACHE_SIZE) + break; + entry++; + } + *crc = ~(u32)0; + *crc = btrfs_csum_data(root, addr + start_offset, *crc, + PAGE_CACHE_SIZE - start_offset); + kunmap(page); + + btrfs_csum_final(*crc, (char *)crc); + crc++; + + bytes += PAGE_CACHE_SIZE; + + ClearPageChecked(page); + set_page_extent_mapped(page); + SetPageUptodate(page); + set_page_dirty(page); + + /* + * We need to release our reference we got for grab_cache_page, + * except for the first page which will hold our checksums, we + * do that below. + */ + if (index != 0) { + unlock_page(page); + page_cache_release(page); + } + + page_cache_release(page); + + index++; + } while (node); + + /* Write out the bitmaps */ + list_for_each_safe(pos, n, &bitmap_list) { + void *addr; + struct btrfs_free_space *entry = + list_entry(pos, struct btrfs_free_space, list); + + page = find_get_page(inode->i_mapping, index); + + addr = kmap(page); + memcpy(addr, entry->bitmap, PAGE_CACHE_SIZE); + *crc = ~(u32)0; + *crc = btrfs_csum_data(root, addr, *crc, PAGE_CACHE_SIZE); + kunmap(page); + btrfs_csum_final(*crc, (char *)crc); + crc++; + bytes += PAGE_CACHE_SIZE; + + ClearPageChecked(page); + set_page_extent_mapped(page); + SetPageUptodate(page); + set_page_dirty(page); + unlock_page(page); + page_cache_release(page); + page_cache_release(page); + list_del_init(&entry->list); + index++; + } + + /* Zero out the rest of the pages just to make sure */ + while (index <= last_index) { + void *addr; + + page = find_get_page(inode->i_mapping, index); + + addr = kmap(page); + memset(addr, 0, PAGE_CACHE_SIZE); + kunmap(page); + ClearPageChecked(page); + set_page_extent_mapped(page); + SetPageUptodate(page); + set_page_dirty(page); + unlock_page(page); + page_cache_release(page); + page_cache_release(page); + bytes += PAGE_CACHE_SIZE; + index++; + } + + btrfs_set_extent_delalloc(inode, 0, bytes - 1, &cached_state); + + /* Write the checksums and trans id to the first page */ + { + void *addr; + u64 *gen; + + page = find_get_page(inode->i_mapping, 0); + + addr = kmap(page); + memcpy(addr, checksums, sizeof(u32) * num_checksums); + gen = addr + (sizeof(u32) * num_checksums); + *gen = trans->transid; + kunmap(page); + ClearPageChecked(page); + set_page_extent_mapped(page); + SetPageUptodate(page); + set_page_dirty(page); + unlock_page(page); + page_cache_release(page); + page_cache_release(page); + } + BTRFS_I(inode)->generation = trans->transid; + + unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0, + i_size_read(inode) - 1, &cached_state, GFP_NOFS); + + filemap_write_and_wait(inode->i_mapping); + + key.objectid = BTRFS_FREE_SPACE_OBJECTID; + key.offset = block_group->key.objectid; + key.type = 0; + + ret = btrfs_search_slot(trans, root, &key, path, 1, 1); + if (ret < 0) { + ret = 0; + clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, bytes - 1, + EXTENT_DIRTY | EXTENT_DELALLOC | + EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS); + goto out_free; + } + leaf = path->nodes[0]; + if (ret > 0) { + struct btrfs_key found_key; + BUG_ON(!path->slots[0]); + path->slots[0]--; + btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); + if (found_key.objectid != BTRFS_FREE_SPACE_OBJECTID || + found_key.offset != block_group->key.objectid) { + ret = 0; + clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, bytes - 1, + EXTENT_DIRTY | EXTENT_DELALLOC | + EXTENT_DO_ACCOUNTING, 0, 0, NULL, + GFP_NOFS); + btrfs_release_path(root, path); + goto out_free; + } + } + header = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_free_space_header); + btrfs_set_free_space_entries(leaf, header, entries); + btrfs_set_free_space_bitmaps(leaf, header, bitmaps); + btrfs_set_free_space_generation(leaf, header, trans->transid); + btrfs_mark_buffer_dirty(leaf); + btrfs_release_path(root, path); + + ret = 1; + +out_free: + if (ret == 0) { + invalidate_inode_pages2_range(inode->i_mapping, 0, index); + spin_lock(&block_group->lock); + block_group->disk_cache_state = BTRFS_DC_ERROR; + spin_unlock(&block_group->lock); + BTRFS_I(inode)->generation = 0; + } + kfree(checksums); + btrfs_update_inode(trans, root, inode); + iput(inode); + return ret; +} + static inline unsigned long offset_to_bit(u64 bitmap_start, u64 sectorsize, u64 offset) { diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h index 890a8e7..e49ca5c 100644 --- a/fs/btrfs/free-space-cache.h +++ b/fs/btrfs/free-space-cache.h @@ -27,6 +27,24 @@ struct btrfs_free_space { struct list_head list; }; +struct inode *lookup_free_space_inode(struct btrfs_root *root, + struct btrfs_block_group_cache + *block_group, struct btrfs_path *path); +int create_free_space_inode(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_block_group_cache *block_group, + struct btrfs_path *path); + +int btrfs_truncate_free_space_cache(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_path *path, + struct inode *inode); +int load_free_space_cache(struct btrfs_fs_info *fs_info, + struct btrfs_block_group_cache *block_group); +int btrfs_write_out_cache(struct btrfs_root *root, + struct btrfs_trans_handle *trans, + struct btrfs_block_group_cache *block_group, + struct btrfs_path *path); int btrfs_add_free_space(struct btrfs_block_group_cache *block_group, u64 bytenr, u64 size); int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 64f99cf..558cac2 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -319,8 +319,6 @@ static noinline int compress_file_range(struct inode *inode, struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; u64 num_bytes; - u64 orig_start; - u64 disk_num_bytes; u64 blocksize = root->sectorsize; u64 actual_end; u64 isize = i_size_read(inode); @@ -335,8 +333,6 @@ static noinline int compress_file_range(struct inode *inode, int i; int will_compress; - orig_start = start; - actual_end = min_t(u64, isize, end + 1); again: will_compress = 0; @@ -371,7 +367,6 @@ again: total_compressed = min(total_compressed, max_uncompressed); num_bytes = (end - start + blocksize) & ~(blocksize - 1); num_bytes = max(blocksize, num_bytes); - disk_num_bytes = num_bytes; total_in = 0; ret = 0; @@ -467,7 +462,6 @@ again: if (total_compressed >= total_in) { will_compress = 0; } else { - disk_num_bytes = total_compressed; num_bytes = total_in; } } @@ -757,20 +751,17 @@ static noinline int cow_file_range(struct inode *inode, u64 disk_num_bytes; u64 cur_alloc_size; u64 blocksize = root->sectorsize; - u64 actual_end; - u64 isize = i_size_read(inode); struct btrfs_key ins; struct extent_map *em; struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; int ret = 0; + BUG_ON(root == root->fs_info->tree_root); trans = btrfs_join_transaction(root, 1); BUG_ON(!trans); btrfs_set_trans_block_group(trans, inode); trans->block_rsv = &root->fs_info->delalloc_block_rsv; - actual_end = min_t(u64, isize, end + 1); - num_bytes = (end - start + blocksize) & ~(blocksize - 1); num_bytes = max(blocksize, num_bytes); disk_num_bytes = num_bytes; @@ -1035,10 +1026,16 @@ static noinline int run_delalloc_nocow(struct inode *inode, int type; int nocow; int check_prev = 1; + bool nolock = false; path = btrfs_alloc_path(); BUG_ON(!path); - trans = btrfs_join_transaction(root, 1); + if (root == root->fs_info->tree_root) { + nolock = true; + trans = btrfs_join_transaction_nolock(root, 1); + } else { + trans = btrfs_join_transaction(root, 1); + } BUG_ON(!trans); cow_start = (u64)-1; @@ -1211,8 +1208,13 @@ out_check: BUG_ON(ret); } - ret = btrfs_end_transaction(trans, root); - BUG_ON(ret); + if (nolock) { + ret = btrfs_end_transaction_nolock(trans, root); + BUG_ON(ret); + } else { + ret = btrfs_end_transaction(trans, root); + BUG_ON(ret); + } btrfs_free_path(path); return 0; } @@ -1289,6 +1291,8 @@ static int btrfs_set_bit_hook(struct inode *inode, if (!(state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; + int do_list = (root->root_key.objectid != + BTRFS_ROOT_TREE_OBJECTID); if (*bits & EXTENT_FIRST_DELALLOC) *bits &= ~EXTENT_FIRST_DELALLOC; @@ -1298,7 +1302,7 @@ static int btrfs_set_bit_hook(struct inode *inode, spin_lock(&root->fs_info->delalloc_lock); BTRFS_I(inode)->delalloc_bytes += len; root->fs_info->delalloc_bytes += len; - if (list_empty(&BTRFS_I(inode)->delalloc_inodes)) { + if (do_list && list_empty(&BTRFS_I(inode)->delalloc_inodes)) { list_add_tail(&BTRFS_I(inode)->delalloc_inodes, &root->fs_info->delalloc_inodes); } @@ -1321,6 +1325,8 @@ static int btrfs_clear_bit_hook(struct inode *inode, if ((state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; + int do_list = (root->root_key.objectid != + BTRFS_ROOT_TREE_OBJECTID); if (*bits & EXTENT_FIRST_DELALLOC) *bits &= ~EXTENT_FIRST_DELALLOC; @@ -1330,14 +1336,15 @@ static int btrfs_clear_bit_hook(struct inode *inode, if (*bits & EXTENT_DO_ACCOUNTING) btrfs_delalloc_release_metadata(inode, len); - if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) + if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID + && do_list) btrfs_free_reserved_data_space(inode, len); spin_lock(&root->fs_info->delalloc_lock); root->fs_info->delalloc_bytes -= len; BTRFS_I(inode)->delalloc_bytes -= len; - if (BTRFS_I(inode)->delalloc_bytes == 0 && + if (do_list && BTRFS_I(inode)->delalloc_bytes == 0 && !list_empty(&BTRFS_I(inode)->delalloc_inodes)) { list_del_init(&BTRFS_I(inode)->delalloc_inodes); } @@ -1372,7 +1379,7 @@ int btrfs_merge_bio_hook(struct page *page, unsigned long offset, if (map_length < length + size) return 1; - return 0; + return ret; } /* @@ -1426,7 +1433,10 @@ static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); + if (root == root->fs_info->tree_root) + ret = btrfs_bio_wq_end_io(root->fs_info, bio, 2); + else + ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); BUG_ON(ret); if (!(rw & REQ_WRITE)) { @@ -1662,6 +1672,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) struct extent_state *cached_state = NULL; int compressed = 0; int ret; + bool nolock = false; ret = btrfs_dec_test_ordered_pending(inode, &ordered_extent, start, end - start + 1); @@ -1669,11 +1680,17 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) return 0; BUG_ON(!ordered_extent); + nolock = (root == root->fs_info->tree_root); + if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) { BUG_ON(!list_empty(&ordered_extent->list)); ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent); if (!ret) { - trans = btrfs_join_transaction(root, 1); + if (nolock) + trans = btrfs_join_transaction_nolock(root, 1); + else + trans = btrfs_join_transaction(root, 1); + BUG_ON(!trans); btrfs_set_trans_block_group(trans, inode); trans->block_rsv = &root->fs_info->delalloc_block_rsv; ret = btrfs_update_inode(trans, root, inode); @@ -1686,7 +1703,10 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) ordered_extent->file_offset + ordered_extent->len - 1, 0, &cached_state, GFP_NOFS); - trans = btrfs_join_transaction(root, 1); + if (nolock) + trans = btrfs_join_transaction_nolock(root, 1); + else + trans = btrfs_join_transaction(root, 1); btrfs_set_trans_block_group(trans, inode); trans->block_rsv = &root->fs_info->delalloc_block_rsv; @@ -1700,6 +1720,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) ordered_extent->len); BUG_ON(ret); } else { + BUG_ON(root == root->fs_info->tree_root); ret = insert_reserved_file_extent(trans, inode, ordered_extent->file_offset, ordered_extent->start, @@ -1724,9 +1745,15 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) ret = btrfs_update_inode(trans, root, inode); BUG_ON(ret); out: - btrfs_delalloc_release_metadata(inode, ordered_extent->len); - if (trans) - btrfs_end_transaction(trans, root); + if (nolock) { + if (trans) + btrfs_end_transaction_nolock(trans, root); + } else { + btrfs_delalloc_release_metadata(inode, ordered_extent->len); + if (trans) + btrfs_end_transaction(trans, root); + } + /* once for us */ btrfs_put_ordered_extent(ordered_extent); /* once for the tree */ @@ -2237,7 +2264,6 @@ void btrfs_orphan_cleanup(struct btrfs_root *root) { struct btrfs_path *path; struct extent_buffer *leaf; - struct btrfs_item *item; struct btrfs_key key, found_key; struct btrfs_trans_handle *trans; struct inode *inode; @@ -2275,7 +2301,6 @@ void btrfs_orphan_cleanup(struct btrfs_root *root) /* pull out the item */ leaf = path->nodes[0]; - item = btrfs_item_nr(leaf, path->slots[0]); btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); /* make sure the item matches what we want */ @@ -2651,7 +2676,8 @@ int btrfs_unlink_inode(struct btrfs_trans_handle *trans, ret = btrfs_del_dir_entries_in_log(trans, root, name, name_len, dir, index); - BUG_ON(ret); + if (ret == -ENOENT) + ret = 0; err: btrfs_free_path(path); if (ret) @@ -2672,8 +2698,8 @@ static int check_path_shared(struct btrfs_root *root, { struct extent_buffer *eb; int level; - int ret; u64 refs = 1; + int uninitialized_var(ret); for (level = 0; level < BTRFS_MAX_LEVEL; level++) { if (!path->nodes[level]) @@ -2686,7 +2712,7 @@ static int check_path_shared(struct btrfs_root *root, if (refs > 1) return 1; } - return 0; + return ret; /* XXX callers? */ } /* @@ -3196,7 +3222,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY); - if (root->ref_cows) + if (root->ref_cows || root == root->fs_info->tree_root) btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0); path = btrfs_alloc_path(); @@ -3344,7 +3370,8 @@ delete: } else { break; } - if (found_extent && root->ref_cows) { + if (found_extent && (root->ref_cows || + root == root->fs_info->tree_root)) { btrfs_set_path_blocking(path); ret = btrfs_free_extent(trans, root, extent_start, extent_num_bytes, 0, @@ -3675,7 +3702,8 @@ void btrfs_evict_inode(struct inode *inode) int ret; truncate_inode_pages(&inode->i_data, 0); - if (inode->i_nlink && btrfs_root_refs(&root->root_item) != 0) + if (inode->i_nlink && (btrfs_root_refs(&root->root_item) != 0 || + root == root->fs_info->tree_root)) goto no_delete; if (is_bad_inode(inode)) { @@ -3888,7 +3916,14 @@ static void inode_tree_del(struct inode *inode) } spin_unlock(&root->inode_lock); - if (empty && btrfs_root_refs(&root->root_item) == 0) { + /* + * Free space cache has inodes in the tree root, but the tree root has a + * root_refs of 0, so this could end up dropping the tree root as a + * snapshot, so we need the extra !root->fs_info->tree_root check to + * make sure we don't drop it. + */ + if (empty && btrfs_root_refs(&root->root_item) == 0 && + root != root->fs_info->tree_root) { synchronize_srcu(&root->fs_info->subvol_srcu); spin_lock(&root->inode_lock); empty = RB_EMPTY_ROOT(&root->inode_tree); @@ -4282,14 +4317,24 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; int ret = 0; + bool nolock = false; if (BTRFS_I(inode)->dummy_inode) return 0; + smp_mb(); + nolock = (root->fs_info->closing && root == root->fs_info->tree_root); + if (wbc->sync_mode == WB_SYNC_ALL) { - trans = btrfs_join_transaction(root, 1); + if (nolock) + trans = btrfs_join_transaction_nolock(root, 1); + else + trans = btrfs_join_transaction(root, 1); btrfs_set_trans_block_group(trans, inode); - ret = btrfs_commit_transaction(trans, root); + if (nolock) + ret = btrfs_end_transaction_nolock(trans, root); + else + ret = btrfs_commit_transaction(trans, root); } return ret; } @@ -5645,7 +5690,6 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode, struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_dio_private *dip; struct bio_vec *bvec = bio->bi_io_vec; - u64 start; int skip_sum; int write = rw & REQ_WRITE; int ret = 0; @@ -5671,7 +5715,6 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode, dip->inode = inode; dip->logical_offset = file_offset; - start = dip->logical_offset; dip->bytes = 0; do { dip->bytes += bvec->bv_len; @@ -6308,6 +6351,21 @@ void btrfs_destroy_inode(struct inode *inode) spin_unlock(&root->fs_info->ordered_extent_lock); } + if (root == root->fs_info->tree_root) { + struct btrfs_block_group_cache *block_group; + + block_group = btrfs_lookup_block_group(root->fs_info, + BTRFS_I(inode)->block_group); + if (block_group && block_group->inode == inode) { + spin_lock(&block_group->lock); + block_group->inode = NULL; + spin_unlock(&block_group->lock); + btrfs_put_block_group(block_group); + } else if (block_group) { + btrfs_put_block_group(block_group); + } + } + spin_lock(&root->orphan_lock); if (!list_empty(&BTRFS_I(inode)->i_orphan)) { printk(KERN_INFO "BTRFS: inode %lu still on the orphan list\n", @@ -6340,7 +6398,8 @@ int btrfs_drop_inode(struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; - if (btrfs_root_refs(&root->root_item) == 0) + if (btrfs_root_refs(&root->root_item) == 0 && + root != root->fs_info->tree_root) return 1; else return generic_drop_inode(inode); @@ -6609,7 +6668,8 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput) return 0; } -int btrfs_start_one_delalloc_inode(struct btrfs_root *root, int delay_iput) +int btrfs_start_one_delalloc_inode(struct btrfs_root *root, int delay_iput, + int sync) { struct btrfs_inode *binode; struct inode *inode = NULL; @@ -6631,7 +6691,26 @@ int btrfs_start_one_delalloc_inode(struct btrfs_root *root, int delay_iput) spin_unlock(&root->fs_info->delalloc_lock); if (inode) { - write_inode_now(inode, 0); + if (sync) { + filemap_write_and_wait(inode->i_mapping); + /* + * We have to do this because compression doesn't + * actually set PG_writeback until it submits the pages + * for IO, which happens in an async thread, so we could + * race and not actually wait for any writeback pages + * because they've not been submitted yet. Technically + * this could still be the case for the ordered stuff + * since the async thread may not have started to do its + * work yet. If this becomes the case then we need to + * figure out a way to make sure that in writepage we + * wait for any async pages to be submitted before + * returning so that fdatawait does what its supposed to + * do. + */ + btrfs_wait_ordered_range(inode, 0, (u64)-1); + } else { + filemap_flush(inode->i_mapping); + } if (delay_iput) btrfs_add_delayed_iput(inode); else @@ -6757,27 +6836,33 @@ out_unlock: return err; } -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) +static 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_trans_handle *trans; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 cur_offset = start; int ret = 0; + bool own_trans = true; + if (trans) + own_trans = false; while (num_bytes > 0) { - trans = btrfs_start_transaction(root, 3); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - break; + if (own_trans) { + trans = btrfs_start_transaction(root, 3); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + break; + } } ret = btrfs_reserve_extent(trans, root, num_bytes, min_size, 0, *alloc_hint, (u64)-1, &ins, 1); if (ret) { - btrfs_end_transaction(trans, root); + if (own_trans) + btrfs_end_transaction(trans, root); break; } @@ -6810,11 +6895,30 @@ int btrfs_prealloc_file_range(struct inode *inode, int mode, ret = btrfs_update_inode(trans, root, inode); BUG_ON(ret); - btrfs_end_transaction(trans, root); + if (own_trans) + btrfs_end_transaction(trans, root); } return ret; } +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) +{ + return __btrfs_prealloc_file_range(inode, mode, start, num_bytes, + min_size, actual_len, alloc_hint, + NULL); +} + +int btrfs_prealloc_file_range_trans(struct inode *inode, + struct btrfs_trans_handle *trans, int mode, + u64 start, u64 num_bytes, u64 min_size, + loff_t actual_len, u64 *alloc_hint) +{ + return __btrfs_prealloc_file_range(inode, mode, start, num_bytes, + min_size, actual_len, alloc_hint, trans); +} + static long btrfs_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len) { diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 9254b3d..463d91b 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -224,7 +224,8 @@ static int btrfs_ioctl_getversion(struct file *file, int __user *arg) static noinline int create_subvol(struct btrfs_root *root, struct dentry *dentry, - char *name, int namelen) + char *name, int namelen, + u64 *async_transid) { struct btrfs_trans_handle *trans; struct btrfs_key key; @@ -338,13 +339,19 @@ static noinline int create_subvol(struct btrfs_root *root, d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry)); fail: - err = btrfs_commit_transaction(trans, root); + if (async_transid) { + *async_transid = trans->transid; + err = btrfs_commit_transaction_async(trans, root, 1); + } else { + err = btrfs_commit_transaction(trans, root); + } if (err && !ret) ret = err; return ret; } -static int create_snapshot(struct btrfs_root *root, struct dentry *dentry) +static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, + char *name, int namelen, u64 *async_transid) { struct inode *inode; struct btrfs_pending_snapshot *pending_snapshot; @@ -373,7 +380,14 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry) list_add(&pending_snapshot->list, &trans->transaction->pending_snapshots); - ret = btrfs_commit_transaction(trans, root->fs_info->extent_root); + if (async_transid) { + *async_transid = trans->transid; + ret = btrfs_commit_transaction_async(trans, + root->fs_info->extent_root, 1); + } else { + ret = btrfs_commit_transaction(trans, + root->fs_info->extent_root); + } BUG_ON(ret); ret = pending_snapshot->error; @@ -395,6 +409,76 @@ fail: return ret; } +/* copy of check_sticky in fs/namei.c() +* It's inline, so penalty for filesystems that don't use sticky bit is +* minimal. +*/ +static inline int btrfs_check_sticky(struct inode *dir, struct inode *inode) +{ + uid_t fsuid = current_fsuid(); + + if (!(dir->i_mode & S_ISVTX)) + return 0; + if (inode->i_uid == fsuid) + return 0; + if (dir->i_uid == fsuid) + return 0; + return !capable(CAP_FOWNER); +} + +/* copy of may_delete in fs/namei.c() + * Check whether we can remove a link victim from directory dir, check + * whether the type of victim is right. + * 1. We can't do it if dir is read-only (done in permission()) + * 2. We should have write and exec permissions on dir + * 3. We can't remove anything from append-only dir + * 4. We can't do anything with immutable dir (done in permission()) + * 5. If the sticky bit on dir is set we should either + * a. be owner of dir, or + * b. be owner of victim, or + * c. have CAP_FOWNER capability + * 6. If the victim is append-only or immutable we can't do antyhing with + * links pointing to it. + * 7. If we were asked to remove a directory and victim isn't one - ENOTDIR. + * 8. If we were asked to remove a non-directory and victim isn't one - EISDIR. + * 9. We can't remove a root or mountpoint. + * 10. We don't allow removal of NFS sillyrenamed files; it's handled by + * nfs_async_unlink(). + */ + +static int btrfs_may_delete(struct inode *dir,struct dentry *victim,int isdir) +{ + int error; + + if (!victim->d_inode) + return -ENOENT; + + BUG_ON(victim->d_parent->d_inode != dir); + audit_inode_child(victim, dir); + + error = inode_permission(dir, MAY_WRITE | MAY_EXEC); + if (error) + return error; + if (IS_APPEND(dir)) + return -EPERM; + if (btrfs_check_sticky(dir, victim->d_inode)|| + IS_APPEND(victim->d_inode)|| + IS_IMMUTABLE(victim->d_inode) || IS_SWAPFILE(victim->d_inode)) + return -EPERM; + if (isdir) { + if (!S_ISDIR(victim->d_inode->i_mode)) + return -ENOTDIR; + if (IS_ROOT(victim)) + return -EBUSY; + } else if (S_ISDIR(victim->d_inode->i_mode)) + return -EISDIR; + if (IS_DEADDIR(dir)) + return -ENOENT; + if (victim->d_flags & DCACHE_NFSFS_RENAMED) + return -EBUSY; + return 0; +} + /* copy of may_create in fs/namei.c() */ static inline int btrfs_may_create(struct inode *dir, struct dentry *child) { @@ -412,7 +496,8 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child) */ static noinline int btrfs_mksubvol(struct path *parent, char *name, int namelen, - struct btrfs_root *snap_src) + struct btrfs_root *snap_src, + u64 *async_transid) { struct inode *dir = parent->dentry->d_inode; struct dentry *dentry; @@ -443,10 +528,11 @@ static noinline int btrfs_mksubvol(struct path *parent, goto out_up_read; if (snap_src) { - error = create_snapshot(snap_src, dentry); + error = create_snapshot(snap_src, dentry, + name, namelen, async_transid); } else { error = create_subvol(BTRFS_I(dir)->root, dentry, - name, namelen); + name, namelen, async_transid); } if (!error) fsnotify_mkdir(dir, dentry); @@ -708,7 +794,6 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root, char *sizestr; char *devstr = NULL; int ret = 0; - int namelen; int mod = 0; if (root->fs_info->sb->s_flags & MS_RDONLY) @@ -722,7 +807,6 @@ static noinline int btrfs_ioctl_resize(struct btrfs_root *root, return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - namelen = strlen(vol_args->name); mutex_lock(&root->fs_info->volume_mutex); sizestr = vol_args->name; @@ -801,11 +885,13 @@ out_unlock: return ret; } -static noinline int btrfs_ioctl_snap_create(struct file *file, - void __user *arg, int subvol) +static noinline int btrfs_ioctl_snap_create_transid(struct file *file, + char *name, + unsigned long fd, + int subvol, + u64 *transid) { struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; - struct btrfs_ioctl_vol_args *vol_args; struct file *src_file; int namelen; int ret = 0; @@ -813,23 +899,18 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, if (root->fs_info->sb->s_flags & MS_RDONLY) return -EROFS; - vol_args = memdup_user(arg, sizeof(*vol_args)); - if (IS_ERR(vol_args)) - return PTR_ERR(vol_args); - - vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - namelen = strlen(vol_args->name); - if (strchr(vol_args->name, '/')) { + namelen = strlen(name); + if (strchr(name, '/')) { ret = -EINVAL; goto out; } if (subvol) { - ret = btrfs_mksubvol(&file->f_path, vol_args->name, namelen, - NULL); + ret = btrfs_mksubvol(&file->f_path, name, namelen, + NULL, transid); } else { struct inode *src_inode; - src_file = fget(vol_args->fd); + src_file = fget(fd); if (!src_file) { ret = -EINVAL; goto out; @@ -843,12 +924,56 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, fput(src_file); goto out; } - ret = btrfs_mksubvol(&file->f_path, vol_args->name, namelen, - BTRFS_I(src_inode)->root); + ret = btrfs_mksubvol(&file->f_path, name, namelen, + BTRFS_I(src_inode)->root, + transid); fput(src_file); } out: + return ret; +} + +static noinline int btrfs_ioctl_snap_create(struct file *file, + void __user *arg, int subvol, + int async) +{ + struct btrfs_ioctl_vol_args *vol_args = NULL; + struct btrfs_ioctl_async_vol_args *async_vol_args = NULL; + char *name; + u64 fd; + u64 transid = 0; + int ret; + + if (async) { + async_vol_args = memdup_user(arg, sizeof(*async_vol_args)); + if (IS_ERR(async_vol_args)) + return PTR_ERR(async_vol_args); + + name = async_vol_args->name; + fd = async_vol_args->fd; + async_vol_args->name[BTRFS_SNAPSHOT_NAME_MAX] = '\0'; + } else { + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); + name = vol_args->name; + fd = vol_args->fd; + vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; + } + + ret = btrfs_ioctl_snap_create_transid(file, name, fd, + subvol, &transid); + + if (!ret && async) { + if (copy_to_user(arg + + offsetof(struct btrfs_ioctl_async_vol_args, + transid), &transid, sizeof(transid))) + return -EFAULT; + } + kfree(vol_args); + kfree(async_vol_args); + return ret; } @@ -1073,14 +1198,10 @@ static noinline int btrfs_ioctl_tree_search(struct file *file, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - args = kmalloc(sizeof(*args), GFP_KERNEL); - if (!args) - return -ENOMEM; + args = memdup_user(argp, sizeof(*args)); + if (IS_ERR(args)) + return PTR_ERR(args); - if (copy_from_user(args, argp, sizeof(*args))) { - kfree(args); - return -EFAULT; - } inode = fdentry(file)->d_inode; ret = search_ioctl(inode, args); if (ret == 0 && copy_to_user(argp, args, sizeof(*args))) @@ -1188,14 +1309,10 @@ static noinline int btrfs_ioctl_ino_lookup(struct file *file, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - args = kmalloc(sizeof(*args), GFP_KERNEL); - if (!args) - return -ENOMEM; + args = memdup_user(argp, sizeof(*args)); + if (IS_ERR(args)) + return PTR_ERR(args); - if (copy_from_user(args, argp, sizeof(*args))) { - kfree(args); - return -EFAULT; - } inode = fdentry(file)->d_inode; if (args->treeid == 0) @@ -1227,9 +1344,6 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, int ret; int err = 0; - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) return PTR_ERR(vol_args); @@ -1259,13 +1373,51 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, } inode = dentry->d_inode; + dest = BTRFS_I(inode)->root; + if (!capable(CAP_SYS_ADMIN)){ + /* + * Regular user. Only allow this with a special mount + * option, when the user has write+exec access to the + * subvol root, and when rmdir(2) would have been + * allowed. + * + * Note that this is _not_ check that the subvol is + * empty or doesn't contain data that we wouldn't + * otherwise be able to delete. + * + * Users who want to delete empty subvols should try + * rmdir(2). + */ + err = -EPERM; + if (!btrfs_test_opt(root, USER_SUBVOL_RM_ALLOWED)) + goto out_dput; + + /* + * Do not allow deletion if the parent dir is the same + * as the dir to be deleted. That means the ioctl + * must be called on the dentry referencing the root + * of the subvol, not a random directory contained + * within it. + */ + err = -EINVAL; + if (root == dest) + goto out_dput; + + err = inode_permission(inode, MAY_WRITE | MAY_EXEC); + if (err) + goto out_dput; + + /* check if subvolume may be deleted by a non-root user */ + err = btrfs_may_delete(dir, dentry, 1); + if (err) + goto out_dput; + } + if (inode->i_ino != BTRFS_FIRST_FREE_OBJECTID) { err = -EINVAL; goto out_dput; } - dest = BTRFS_I(inode)->root; - mutex_lock(&inode->i_mutex); err = d_invalidate(dentry); if (err) @@ -1304,7 +1456,7 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, BUG_ON(ret); } - ret = btrfs_commit_transaction(trans, root); + ret = btrfs_end_transaction(trans, root); BUG_ON(ret); inode->i_flags |= S_DEAD; out_up_write: @@ -1502,11 +1654,11 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, path->reada = 2; if (inode < src) { - mutex_lock(&inode->i_mutex); - mutex_lock(&src->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&src->i_mutex, I_MUTEX_CHILD); } else { - mutex_lock(&src->i_mutex); - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&src->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); } /* determine range to clone */ @@ -1530,13 +1682,15 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, while (1) { struct btrfs_ordered_extent *ordered; lock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS); - ordered = btrfs_lookup_first_ordered_extent(inode, off+len); - if (BTRFS_I(src)->delalloc_bytes == 0 && !ordered) + ordered = btrfs_lookup_first_ordered_extent(src, off+len); + if (!ordered && + !test_range_bit(&BTRFS_I(src)->io_tree, off, off+len, + EXTENT_DELALLOC, 0, NULL)) break; unlock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS); if (ordered) btrfs_put_ordered_extent(ordered); - btrfs_wait_ordered_range(src, off, off+len); + btrfs_wait_ordered_range(src, off, len); } /* clone data */ @@ -1605,7 +1759,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, } btrfs_release_path(root, path); - if (key.offset + datal < off || + if (key.offset + datal <= off || key.offset >= off+len) goto next; @@ -1879,6 +2033,22 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) return 0; } +static void get_block_group_info(struct list_head *groups_list, + struct btrfs_ioctl_space_info *space) +{ + struct btrfs_block_group_cache *block_group; + + space->total_bytes = 0; + space->used_bytes = 0; + space->flags = 0; + list_for_each_entry(block_group, groups_list, list) { + space->flags = block_group->flags; + space->total_bytes += block_group->key.offset; + space->used_bytes += + btrfs_block_group_used(&block_group->item); + } +} + long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) { struct btrfs_ioctl_space_args space_args; @@ -1887,27 +2057,56 @@ long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) struct btrfs_ioctl_space_info *dest_orig; struct btrfs_ioctl_space_info *user_dest; struct btrfs_space_info *info; + u64 types[] = {BTRFS_BLOCK_GROUP_DATA, + BTRFS_BLOCK_GROUP_SYSTEM, + BTRFS_BLOCK_GROUP_METADATA, + BTRFS_BLOCK_GROUP_DATA | BTRFS_BLOCK_GROUP_METADATA}; + int num_types = 4; int alloc_size; int ret = 0; int slot_count = 0; + int i, c; if (copy_from_user(&space_args, (struct btrfs_ioctl_space_args __user *)arg, sizeof(space_args))) return -EFAULT; - /* first we count slots */ - rcu_read_lock(); - list_for_each_entry_rcu(info, &root->fs_info->space_info, list) - slot_count++; - rcu_read_unlock(); + for (i = 0; i < num_types; i++) { + struct btrfs_space_info *tmp; + + info = NULL; + rcu_read_lock(); + list_for_each_entry_rcu(tmp, &root->fs_info->space_info, + list) { + if (tmp->flags == types[i]) { + info = tmp; + break; + } + } + rcu_read_unlock(); + + if (!info) + continue; + + down_read(&info->groups_sem); + for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { + if (!list_empty(&info->block_groups[c])) + slot_count++; + } + up_read(&info->groups_sem); + } /* space_slots == 0 means they are asking for a count */ if (space_args.space_slots == 0) { space_args.total_spaces = slot_count; goto out; } + + slot_count = min_t(int, space_args.space_slots, slot_count); + alloc_size = sizeof(*dest) * slot_count; + /* we generally have at most 6 or so space infos, one for each raid * level. So, a whole page should be more than enough for everyone */ @@ -1921,27 +2120,34 @@ long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg) dest_orig = dest; /* now we have a buffer to copy into */ - rcu_read_lock(); - list_for_each_entry_rcu(info, &root->fs_info->space_info, list) { - /* make sure we don't copy more than we allocated - * in our buffer - */ - if (slot_count == 0) - break; - slot_count--; - - /* make sure userland has enough room in their buffer */ - if (space_args.total_spaces >= space_args.space_slots) - break; + for (i = 0; i < num_types; i++) { + struct btrfs_space_info *tmp; + + info = NULL; + rcu_read_lock(); + list_for_each_entry_rcu(tmp, &root->fs_info->space_info, + list) { + if (tmp->flags == types[i]) { + info = tmp; + break; + } + } + rcu_read_unlock(); - space.flags = info->flags; - space.total_bytes = info->total_bytes; - space.used_bytes = info->bytes_used; - memcpy(dest, &space, sizeof(space)); - dest++; - space_args.total_spaces++; + if (!info) + continue; + down_read(&info->groups_sem); + for (c = 0; c < BTRFS_NR_RAID_TYPES; c++) { + if (!list_empty(&info->block_groups[c])) { + get_block_group_info(&info->block_groups[c], + &space); + memcpy(dest, &space, sizeof(space)); + dest++; + space_args.total_spaces++; + } + } + up_read(&info->groups_sem); } - rcu_read_unlock(); user_dest = (struct btrfs_ioctl_space_info *) (arg + sizeof(struct btrfs_ioctl_space_args)); @@ -1984,6 +2190,36 @@ long btrfs_ioctl_trans_end(struct file *file) return 0; } +static noinline long btrfs_ioctl_start_sync(struct file *file, void __user *argp) +{ + struct btrfs_root *root = BTRFS_I(file->f_dentry->d_inode)->root; + struct btrfs_trans_handle *trans; + u64 transid; + + trans = btrfs_start_transaction(root, 0); + transid = trans->transid; + btrfs_commit_transaction_async(trans, root, 0); + + if (argp) + if (copy_to_user(argp, &transid, sizeof(transid))) + return -EFAULT; + return 0; +} + +static noinline long btrfs_ioctl_wait_sync(struct file *file, void __user *argp) +{ + struct btrfs_root *root = BTRFS_I(file->f_dentry->d_inode)->root; + u64 transid; + + if (argp) { + if (copy_from_user(&transid, argp, sizeof(transid))) + return -EFAULT; + } else { + transid = 0; /* current trans */ + } + return btrfs_wait_for_commit(root, transid); +} + long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { @@ -1998,9 +2234,11 @@ long btrfs_ioctl(struct file *file, unsigned int case FS_IOC_GETVERSION: return btrfs_ioctl_getversion(file, argp); case BTRFS_IOC_SNAP_CREATE: - return btrfs_ioctl_snap_create(file, argp, 0); + return btrfs_ioctl_snap_create(file, argp, 0, 0); + case BTRFS_IOC_SNAP_CREATE_ASYNC: + return btrfs_ioctl_snap_create(file, argp, 0, 1); case BTRFS_IOC_SUBVOL_CREATE: - return btrfs_ioctl_snap_create(file, argp, 1); + return btrfs_ioctl_snap_create(file, argp, 1, 0); case BTRFS_IOC_SNAP_DESTROY: return btrfs_ioctl_snap_destroy(file, argp); case BTRFS_IOC_DEFAULT_SUBVOL: @@ -2034,6 +2272,10 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_SYNC: btrfs_sync_fs(file->f_dentry->d_sb, 1); return 0; + case BTRFS_IOC_START_SYNC: + return btrfs_ioctl_start_sync(file, argp); + case BTRFS_IOC_WAIT_SYNC: + return btrfs_ioctl_wait_sync(file, argp); } return -ENOTTY; diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h index 424694a..17c99eb 100644 --- a/fs/btrfs/ioctl.h +++ b/fs/btrfs/ioctl.h @@ -22,14 +22,21 @@ #define BTRFS_IOCTL_MAGIC 0x94 #define BTRFS_VOL_NAME_MAX 255 -#define BTRFS_PATH_NAME_MAX 4087 /* this should be 4k */ +#define BTRFS_PATH_NAME_MAX 4087 struct btrfs_ioctl_vol_args { __s64 fd; char name[BTRFS_PATH_NAME_MAX + 1]; }; +#define BTRFS_SNAPSHOT_NAME_MAX 4079 +struct btrfs_ioctl_async_vol_args { + __s64 fd; + __u64 transid; + char name[BTRFS_SNAPSHOT_NAME_MAX + 1]; +}; + #define BTRFS_INO_LOOKUP_PATH_MAX 4080 struct btrfs_ioctl_ino_lookup_args { __u64 treeid; @@ -178,4 +185,8 @@ struct btrfs_ioctl_space_args { #define BTRFS_IOC_DEFAULT_SUBVOL _IOW(BTRFS_IOCTL_MAGIC, 19, u64) #define BTRFS_IOC_SPACE_INFO _IOWR(BTRFS_IOCTL_MAGIC, 20, \ struct btrfs_ioctl_space_args) +#define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64) +#define BTRFS_IOC_WAIT_SYNC _IOW(BTRFS_IOCTL_MAGIC, 22, __u64) +#define BTRFS_IOC_SNAP_CREATE_ASYNC _IOW(BTRFS_IOCTL_MAGIC, 23, \ + struct btrfs_ioctl_async_vol_args) #endif diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index e56c72b..f4621f6 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -526,7 +526,6 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len) { u64 end; u64 orig_end; - u64 wait_end; struct btrfs_ordered_extent *ordered; int found; @@ -537,7 +536,6 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len) if (orig_end > INT_LIMIT(loff_t)) orig_end = INT_LIMIT(loff_t); } - wait_end = orig_end; again: /* start IO across the range first to instantiate any delalloc * extents diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index b37d723..045c9c2 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -29,6 +29,7 @@ #include "locking.h" #include "btrfs_inode.h" #include "async-thread.h" +#include "free-space-cache.h" /* * backref_node, mapping_node and tree_block start with this @@ -178,8 +179,6 @@ struct reloc_control { u64 search_start; u64 extents_found; - int block_rsv_retries; - unsigned int stage:8; unsigned int create_reloc_tree:1; unsigned int merge_reloc_tree:1; @@ -2133,7 +2132,6 @@ int prepare_to_merge(struct reloc_control *rc, int err) LIST_HEAD(reloc_roots); u64 num_bytes = 0; int ret; - int retries = 0; mutex_lock(&root->fs_info->trans_mutex); rc->merging_rsv_size += root->nodesize * (BTRFS_MAX_LEVEL - 1) * 2; @@ -2143,7 +2141,7 @@ again: if (!err) { num_bytes = rc->merging_rsv_size; ret = btrfs_block_rsv_add(NULL, root, rc->block_rsv, - num_bytes, &retries); + num_bytes); if (ret) err = ret; } @@ -2155,7 +2153,6 @@ again: btrfs_end_transaction(trans, rc->extent_root); btrfs_block_rsv_release(rc->extent_root, rc->block_rsv, num_bytes); - retries = 0; goto again; } } @@ -2405,15 +2402,13 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans, num_bytes = calcu_metadata_size(rc, node, 1) * 2; trans->block_rsv = rc->block_rsv; - ret = btrfs_block_rsv_add(trans, root, rc->block_rsv, num_bytes, - &rc->block_rsv_retries); + ret = btrfs_block_rsv_add(trans, root, rc->block_rsv, num_bytes); if (ret) { if (ret == -EAGAIN) rc->commit_transaction = 1; return ret; } - rc->block_rsv_retries = 0; return 0; } @@ -3099,6 +3094,8 @@ static int add_tree_block(struct reloc_control *rc, BUG_ON(item_size != sizeof(struct btrfs_extent_item_v0)); ret = get_ref_objectid_v0(rc, path, extent_key, &ref_owner, NULL); + if (ret < 0) + return ret; BUG_ON(ref_owner >= BTRFS_MAX_LEVEL); level = (int)ref_owner; /* FIXME: get real generation */ @@ -3191,6 +3188,54 @@ static int block_use_full_backref(struct reloc_control *rc, return ret; } +static int delete_block_group_cache(struct btrfs_fs_info *fs_info, + struct inode *inode, u64 ino) +{ + struct btrfs_key key; + struct btrfs_path *path; + struct btrfs_root *root = fs_info->tree_root; + struct btrfs_trans_handle *trans; + unsigned long nr; + int ret = 0; + + if (inode) + goto truncate; + + key.objectid = ino; + key.type = BTRFS_INODE_ITEM_KEY; + key.offset = 0; + + inode = btrfs_iget(fs_info->sb, &key, root, NULL); + if (!inode || IS_ERR(inode) || is_bad_inode(inode)) { + if (inode && !IS_ERR(inode)) + iput(inode); + return -ENOENT; + } + +truncate: + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto out; + } + + trans = btrfs_join_transaction(root, 0); + if (IS_ERR(trans)) { + btrfs_free_path(path); + goto out; + } + + ret = btrfs_truncate_free_space_cache(root, trans, path, inode); + + btrfs_free_path(path); + nr = trans->blocks_used; + btrfs_end_transaction(trans, root); + btrfs_btree_balance_dirty(root, nr); +out: + iput(inode); + return ret; +} + /* * helper to add tree blocks for backref of type BTRFS_EXTENT_DATA_REF_KEY * this function scans fs tree to find blocks reference the data extent @@ -3217,15 +3262,27 @@ static int find_data_references(struct reloc_control *rc, int counted; int ret; - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - ref_root = btrfs_extent_data_ref_root(leaf, ref); ref_objectid = btrfs_extent_data_ref_objectid(leaf, ref); ref_offset = btrfs_extent_data_ref_offset(leaf, ref); ref_count = btrfs_extent_data_ref_count(leaf, ref); + /* + * This is an extent belonging to the free space cache, lets just delete + * it and redo the search. + */ + if (ref_root == BTRFS_ROOT_TREE_OBJECTID) { + ret = delete_block_group_cache(rc->extent_root->fs_info, + NULL, ref_objectid); + if (ret != -ENOENT) + return ret; + ret = 0; + } + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + root = read_fs_root(rc->extent_root->fs_info, ref_root); if (IS_ERR(root)) { err = PTR_ERR(root); @@ -3554,8 +3611,7 @@ int prepare_to_relocate(struct reloc_control *rc) * is no reservation in transaction handle. */ ret = btrfs_block_rsv_add(NULL, rc->extent_root, rc->block_rsv, - rc->extent_root->nodesize * 256, - &rc->block_rsv_retries); + rc->extent_root->nodesize * 256); if (ret) return ret; @@ -3567,7 +3623,6 @@ int prepare_to_relocate(struct reloc_control *rc) rc->extents_found = 0; rc->nodes_relocated = 0; rc->merging_rsv_size = 0; - rc->block_rsv_retries = 0; rc->create_reloc_tree = 1; set_reloc_control(rc); @@ -3860,6 +3915,8 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start) { struct btrfs_fs_info *fs_info = extent_root->fs_info; struct reloc_control *rc; + struct inode *inode; + struct btrfs_path *path; int ret; int rw = 0; int err = 0; @@ -3882,6 +3939,26 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start) rw = 1; } + path = btrfs_alloc_path(); + if (!path) { + err = -ENOMEM; + goto out; + } + + inode = lookup_free_space_inode(fs_info->tree_root, rc->block_group, + path); + btrfs_free_path(path); + + if (!IS_ERR(inode)) + ret = delete_block_group_cache(fs_info, inode, 0); + else + ret = PTR_ERR(inode); + + if (ret && ret != -ENOENT) { + err = ret; + goto out; + } + rc->data_inode = create_reloc_inode(fs_info, rc->block_group); if (IS_ERR(rc->data_inode)) { err = PTR_ERR(rc->data_inode); @@ -4143,7 +4220,7 @@ int btrfs_reloc_clone_csums(struct inode *inode, u64 file_pos, u64 len) btrfs_add_ordered_sum(inode, ordered, sums); } btrfs_put_ordered_extent(ordered); - return 0; + return ret; } void btrfs_reloc_cow_block(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 2d958be..6a1086e 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -181,7 +181,6 @@ int btrfs_insert_root(struct btrfs_trans_handle *trans, struct btrfs_root int btrfs_find_dead_roots(struct btrfs_root *root, u64 objectid) { struct btrfs_root *dead_root; - struct btrfs_item *item; struct btrfs_root_item *ri; struct btrfs_key key; struct btrfs_key found_key; @@ -214,7 +213,6 @@ again: nritems = btrfs_header_nritems(leaf); slot = path->slots[0]; } - item = btrfs_item_nr(leaf, slot); btrfs_item_key_to_cpu(leaf, &key, slot); if (btrfs_key_type(&key) != BTRFS_ROOT_ITEM_KEY) goto next; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 144f8a5..8299a25 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -61,6 +61,8 @@ static void btrfs_put_super(struct super_block *sb) ret = close_ctree(root); sb->s_fs_info = NULL; + + (void)ret; /* FIXME: need to fix VFS to return error? */ } enum { @@ -68,7 +70,8 @@ enum { Opt_nodatacow, Opt_max_inline, Opt_alloc_start, Opt_nobarrier, Opt_ssd, Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl, Opt_compress, Opt_compress_force, Opt_notreelog, Opt_ratio, Opt_flushoncommit, - Opt_discard, Opt_err, + Opt_discard, Opt_space_cache, Opt_clear_cache, Opt_err, + Opt_user_subvol_rm_allowed, }; static match_table_t tokens = { @@ -92,6 +95,9 @@ static match_table_t tokens = { {Opt_flushoncommit, "flushoncommit"}, {Opt_ratio, "metadata_ratio=%d"}, {Opt_discard, "discard"}, + {Opt_space_cache, "space_cache"}, + {Opt_clear_cache, "clear_cache"}, + {Opt_user_subvol_rm_allowed, "user_subvol_rm_allowed"}, {Opt_err, NULL}, }; @@ -235,6 +241,16 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) case Opt_discard: btrfs_set_opt(info->mount_opt, DISCARD); break; + case Opt_space_cache: + printk(KERN_INFO "btrfs: enabling disk space caching\n"); + btrfs_set_opt(info->mount_opt, SPACE_CACHE); + case Opt_clear_cache: + printk(KERN_INFO "btrfs: force clearing of disk cache\n"); + btrfs_set_opt(info->mount_opt, CLEAR_CACHE); + break; + case Opt_user_subvol_rm_allowed: + btrfs_set_opt(info->mount_opt, USER_SUBVOL_RM_ALLOWED); + break; case Opt_err: printk(KERN_INFO "btrfs: unrecognized mount option " "'%s'\n", p); @@ -380,7 +396,7 @@ static struct dentry *get_default_root(struct super_block *sb, find_root: new_root = btrfs_read_fs_root_no_name(root->fs_info, &location); if (IS_ERR(new_root)) - return ERR_PTR(PTR_ERR(new_root)); + return ERR_CAST(new_root); if (btrfs_root_refs(&new_root->root_item) == 0) return ERR_PTR(-ENOENT); @@ -436,7 +452,6 @@ static int btrfs_fill_super(struct super_block *sb, { struct inode *inode; struct dentry *root_dentry; - struct btrfs_super_block *disk_super; struct btrfs_root *tree_root; struct btrfs_key key; int err; @@ -458,7 +473,6 @@ static int btrfs_fill_super(struct super_block *sb, return PTR_ERR(tree_root); } sb->s_fs_info = tree_root; - disk_super = &tree_root->fs_info->super_copy; key.objectid = BTRFS_FIRST_FREE_OBJECTID; key.type = BTRFS_INODE_ITEM_KEY; @@ -560,8 +574,8 @@ static int btrfs_test_super(struct super_block *s, void *data) * Note: This is based on get_sb_bdev from fs/super.c with a few additions * for multiple device setup. Make sure to keep it in sync. */ -static int btrfs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { struct block_device *bdev = NULL; struct super_block *s; @@ -571,7 +585,6 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, char *subvol_name = NULL; u64 subvol_objectid = 0; int error = 0; - int found = 0; if (!(flags & MS_RDONLY)) mode |= FMODE_WRITE; @@ -580,7 +593,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, &subvol_name, &subvol_objectid, &fs_devices); if (error) - return error; + return ERR_PTR(error); error = btrfs_scan_one_device(dev_name, mode, fs_type, &fs_devices); if (error) @@ -607,7 +620,6 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, goto error_close_devices; } - found = 1; btrfs_close_devices(fs_devices); } else { char b[BDEVNAME_SIZE]; @@ -629,7 +641,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, if (IS_ERR(root)) { error = PTR_ERR(root); deactivate_locked_super(s); - goto error; + goto error_free_subvol_name; } /* if they gave us a subvolume name bind mount into that */ if (strcmp(subvol_name, ".")) { @@ -643,24 +655,21 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags, deactivate_locked_super(s); error = PTR_ERR(new_root); dput(root); - goto error_close_devices; + goto error_free_subvol_name; } if (!new_root->d_inode) { dput(root); dput(new_root); deactivate_locked_super(s); error = -ENXIO; - goto error_close_devices; + goto error_free_subvol_name; } dput(root); root = new_root; } - mnt->mnt_sb = s; - mnt->mnt_root = root; - kfree(subvol_name); - return 0; + return root; error_s: error = PTR_ERR(s); @@ -668,8 +677,7 @@ error_close_devices: btrfs_close_devices(fs_devices); error_free_subvol_name: kfree(subvol_name); -error: - return error; + return ERR_PTR(error); } static int btrfs_remount(struct super_block *sb, int *flags, char *data) @@ -716,18 +724,25 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf) struct list_head *head = &root->fs_info->space_info; struct btrfs_space_info *found; u64 total_used = 0; + u64 total_used_data = 0; int bits = dentry->d_sb->s_blocksize_bits; __be32 *fsid = (__be32 *)root->fs_info->fsid; rcu_read_lock(); - list_for_each_entry_rcu(found, head, list) + list_for_each_entry_rcu(found, head, list) { + if (found->flags & (BTRFS_BLOCK_GROUP_METADATA | + BTRFS_BLOCK_GROUP_SYSTEM)) + total_used_data += found->disk_total; + else + total_used_data += found->disk_used; total_used += found->disk_used; + } rcu_read_unlock(); buf->f_namelen = BTRFS_NAME_LEN; buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits; buf->f_bfree = buf->f_blocks - (total_used >> bits); - buf->f_bavail = buf->f_bfree; + buf->f_bavail = buf->f_blocks - (total_used_data >> bits); buf->f_bsize = dentry->d_sb->s_blocksize; buf->f_type = BTRFS_SUPER_MAGIC; @@ -746,7 +761,7 @@ static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf) static struct file_system_type btrfs_fs_type = { .owner = THIS_MODULE, .name = "btrfs", - .get_sb = btrfs_get_sb, + .mount = btrfs_mount, .kill_sb = kill_anon_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 66e4c66..1fffbc0 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -163,6 +163,7 @@ enum btrfs_trans_type { TRANS_START, TRANS_JOIN, TRANS_USERSPACE, + TRANS_JOIN_NOLOCK, }; static int may_wait_transaction(struct btrfs_root *root, int type) @@ -179,14 +180,14 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, { struct btrfs_trans_handle *h; struct btrfs_transaction *cur_trans; - int retries = 0; int ret; again: h = kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS); if (!h) return ERR_PTR(-ENOMEM); - mutex_lock(&root->fs_info->trans_mutex); + if (type != TRANS_JOIN_NOLOCK) + mutex_lock(&root->fs_info->trans_mutex); if (may_wait_transaction(root, type)) wait_current_trans(root); @@ -195,7 +196,8 @@ again: cur_trans = root->fs_info->running_transaction; cur_trans->use_count++; - mutex_unlock(&root->fs_info->trans_mutex); + if (type != TRANS_JOIN_NOLOCK) + mutex_unlock(&root->fs_info->trans_mutex); h->transid = cur_trans->transid; h->transaction = cur_trans; @@ -212,8 +214,7 @@ again: } if (num_items > 0) { - ret = btrfs_trans_reserve_metadata(h, root, num_items, - &retries); + ret = btrfs_trans_reserve_metadata(h, root, num_items); if (ret == -EAGAIN) { btrfs_commit_transaction(h, root); goto again; @@ -224,9 +225,11 @@ again: } } - mutex_lock(&root->fs_info->trans_mutex); + if (type != TRANS_JOIN_NOLOCK) + mutex_lock(&root->fs_info->trans_mutex); record_root_in_trans(h, root); - mutex_unlock(&root->fs_info->trans_mutex); + if (type != TRANS_JOIN_NOLOCK) + mutex_unlock(&root->fs_info->trans_mutex); if (!current->journal_info && type != TRANS_USERSPACE) current->journal_info = h; @@ -244,6 +247,12 @@ struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root, return start_transaction(root, 0, TRANS_JOIN); } +struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root, + int num_blocks) +{ + return start_transaction(root, 0, TRANS_JOIN_NOLOCK); +} + struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *r, int num_blocks) { @@ -270,6 +279,58 @@ static noinline int wait_for_commit(struct btrfs_root *root, return 0; } +int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid) +{ + struct btrfs_transaction *cur_trans = NULL, *t; + int ret; + + mutex_lock(&root->fs_info->trans_mutex); + + ret = 0; + if (transid) { + if (transid <= root->fs_info->last_trans_committed) + goto out_unlock; + + /* find specified transaction */ + list_for_each_entry(t, &root->fs_info->trans_list, list) { + if (t->transid == transid) { + cur_trans = t; + break; + } + if (t->transid > transid) + break; + } + ret = -EINVAL; + if (!cur_trans) + goto out_unlock; /* bad transid */ + } else { + /* find newest transaction that is committing | committed */ + list_for_each_entry_reverse(t, &root->fs_info->trans_list, + list) { + if (t->in_commit) { + if (t->commit_done) + goto out_unlock; + cur_trans = t; + break; + } + } + if (!cur_trans) + goto out_unlock; /* nothing committing|committed */ + } + + cur_trans->use_count++; + mutex_unlock(&root->fs_info->trans_mutex); + + wait_for_commit(root, cur_trans); + + mutex_lock(&root->fs_info->trans_mutex); + put_transaction(cur_trans); + ret = 0; +out_unlock: + mutex_unlock(&root->fs_info->trans_mutex); + return ret; +} + #if 0 /* * rate limit against the drop_snapshot code. This helps to slow down new @@ -348,7 +409,7 @@ int btrfs_should_end_transaction(struct btrfs_trans_handle *trans, } static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, - struct btrfs_root *root, int throttle) + struct btrfs_root *root, int throttle, int lock) { struct btrfs_transaction *cur_trans = trans->transaction; struct btrfs_fs_info *info = root->fs_info; @@ -376,26 +437,29 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, btrfs_trans_release_metadata(trans, root); - if (!root->fs_info->open_ioctl_trans && + if (lock && !root->fs_info->open_ioctl_trans && should_end_transaction(trans, root)) trans->transaction->blocked = 1; - if (cur_trans->blocked && !cur_trans->in_commit) { + if (lock && cur_trans->blocked && !cur_trans->in_commit) { if (throttle) return btrfs_commit_transaction(trans, root); else wake_up_process(info->transaction_kthread); } - mutex_lock(&info->trans_mutex); + if (lock) + mutex_lock(&info->trans_mutex); WARN_ON(cur_trans != info->running_transaction); WARN_ON(cur_trans->num_writers < 1); cur_trans->num_writers--; + smp_mb(); if (waitqueue_active(&cur_trans->writer_wait)) wake_up(&cur_trans->writer_wait); put_transaction(cur_trans); - mutex_unlock(&info->trans_mutex); + if (lock) + mutex_unlock(&info->trans_mutex); if (current->journal_info == trans) current->journal_info = NULL; @@ -411,13 +475,19 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, int btrfs_end_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root) { - return __btrfs_end_transaction(trans, root, 0); + return __btrfs_end_transaction(trans, root, 0, 1); } int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans, struct btrfs_root *root) { - return __btrfs_end_transaction(trans, root, 1); + return __btrfs_end_transaction(trans, root, 1, 1); +} + +int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans, + struct btrfs_root *root) +{ + return __btrfs_end_transaction(trans, root, 0, 0); } /* @@ -836,7 +906,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, struct extent_buffer *tmp; struct extent_buffer *old; int ret; - int retries = 0; u64 to_reserve = 0; u64 index = 0; u64 objectid; @@ -858,7 +927,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, if (to_reserve > 0) { ret = btrfs_block_rsv_add(trans, root, &pending->block_rsv, - to_reserve, &retries); + to_reserve); if (ret) { pending->error = ret; goto fail; @@ -966,6 +1035,8 @@ static void update_super_roots(struct btrfs_root *root) super->root = root_item->bytenr; super->generation = root_item->generation; super->root_level = root_item->level; + if (super->cache_generation != 0 || btrfs_test_opt(root, SPACE_CACHE)) + super->cache_generation = root_item->generation; } int btrfs_transaction_in_commit(struct btrfs_fs_info *info) @@ -988,11 +1059,127 @@ int btrfs_transaction_blocked(struct btrfs_fs_info *info) return ret; } +/* + * wait for the current transaction commit to start and block subsequent + * transaction joins + */ +static void wait_current_trans_commit_start(struct btrfs_root *root, + struct btrfs_transaction *trans) +{ + DEFINE_WAIT(wait); + + if (trans->in_commit) + return; + + while (1) { + prepare_to_wait(&root->fs_info->transaction_blocked_wait, &wait, + TASK_UNINTERRUPTIBLE); + if (trans->in_commit) { + finish_wait(&root->fs_info->transaction_blocked_wait, + &wait); + break; + } + mutex_unlock(&root->fs_info->trans_mutex); + schedule(); + mutex_lock(&root->fs_info->trans_mutex); + finish_wait(&root->fs_info->transaction_blocked_wait, &wait); + } +} + +/* + * wait for the current transaction to start and then become unblocked. + * caller holds ref. + */ +static void wait_current_trans_commit_start_and_unblock(struct btrfs_root *root, + struct btrfs_transaction *trans) +{ + DEFINE_WAIT(wait); + + if (trans->commit_done || (trans->in_commit && !trans->blocked)) + return; + + while (1) { + prepare_to_wait(&root->fs_info->transaction_wait, &wait, + TASK_UNINTERRUPTIBLE); + if (trans->commit_done || + (trans->in_commit && !trans->blocked)) { + finish_wait(&root->fs_info->transaction_wait, + &wait); + break; + } + mutex_unlock(&root->fs_info->trans_mutex); + schedule(); + mutex_lock(&root->fs_info->trans_mutex); + finish_wait(&root->fs_info->transaction_wait, + &wait); + } +} + +/* + * commit transactions asynchronously. once btrfs_commit_transaction_async + * returns, any subsequent transaction will not be allowed to join. + */ +struct btrfs_async_commit { + struct btrfs_trans_handle *newtrans; + struct btrfs_root *root; + struct delayed_work work; +}; + +static void do_async_commit(struct work_struct *work) +{ + struct btrfs_async_commit *ac = + container_of(work, struct btrfs_async_commit, work.work); + + btrfs_commit_transaction(ac->newtrans, ac->root); + kfree(ac); +} + +int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + int wait_for_unblock) +{ + struct btrfs_async_commit *ac; + struct btrfs_transaction *cur_trans; + + ac = kmalloc(sizeof(*ac), GFP_NOFS); + BUG_ON(!ac); + + INIT_DELAYED_WORK(&ac->work, do_async_commit); + ac->root = root; + ac->newtrans = btrfs_join_transaction(root, 0); + + /* take transaction reference */ + mutex_lock(&root->fs_info->trans_mutex); + cur_trans = trans->transaction; + cur_trans->use_count++; + mutex_unlock(&root->fs_info->trans_mutex); + + btrfs_end_transaction(trans, root); + schedule_delayed_work(&ac->work, 0); + + /* wait for transaction to start and unblock */ + mutex_lock(&root->fs_info->trans_mutex); + if (wait_for_unblock) + wait_current_trans_commit_start_and_unblock(root, cur_trans); + else + wait_current_trans_commit_start(root, cur_trans); + put_transaction(cur_trans); + mutex_unlock(&root->fs_info->trans_mutex); + + return 0; +} + +/* + * btrfs_transaction state sequence: + * in_commit = 0, blocked = 0 (initial) + * in_commit = 1, blocked = 1 + * blocked = 0 + * commit_done = 1 + */ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root) { unsigned long joined = 0; - unsigned long timeout = 1; struct btrfs_transaction *cur_trans; struct btrfs_transaction *prev_trans = NULL; DEFINE_WAIT(wait); @@ -1039,6 +1226,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, trans->transaction->in_commit = 1; trans->transaction->blocked = 1; + wake_up(&root->fs_info->transaction_blocked_wait); + if (cur_trans->list.prev != &root->fs_info->trans_list) { prev_trans = list_entry(cur_trans->list.prev, struct btrfs_transaction, list); @@ -1063,11 +1252,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, snap_pending = 1; WARN_ON(cur_trans != trans->transaction); - if (cur_trans->num_writers > 1) - timeout = MAX_SCHEDULE_TIMEOUT; - else if (should_grow) - timeout = 1; - mutex_unlock(&root->fs_info->trans_mutex); if (flush_on_commit || snap_pending) { @@ -1089,8 +1273,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans, TASK_UNINTERRUPTIBLE); smp_mb(); - if (cur_trans->num_writers > 1 || should_grow) - schedule_timeout(timeout); + if (cur_trans->num_writers > 1) + schedule_timeout(MAX_SCHEDULE_TIMEOUT); + else if (should_grow) + schedule_timeout(1); mutex_lock(&root->fs_info->trans_mutex); finish_wait(&cur_trans->writer_wait, &wait); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index e104986..f104b57 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -87,12 +87,17 @@ static inline void btrfs_set_inode_last_trans(struct btrfs_trans_handle *trans, int btrfs_end_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root); +int btrfs_end_transaction_nolock(struct btrfs_trans_handle *trans, + struct btrfs_root *root); struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root, int num_items); struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root, int num_blocks); +struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root, + int num_blocks); struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *r, int num_blocks); +int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid); int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root); int btrfs_commit_tree_roots(struct btrfs_trans_handle *trans, @@ -104,6 +109,9 @@ int btrfs_defrag_root(struct btrfs_root *root, int cacheonly); int btrfs_clean_old_snapshots(struct btrfs_root *root); int btrfs_commit_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root); +int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans, + struct btrfs_root *root, + int wait_for_unblock); int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans, struct btrfs_root *root); int btrfs_should_end_transaction(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c index f7ac8e0..992ab42 100644 --- a/fs/btrfs/tree-defrag.c +++ b/fs/btrfs/tree-defrag.c @@ -36,7 +36,6 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans, int ret = 0; int wret; int level; - int orig_level; int is_extent = 0; int next_key_ret = 0; u64 last_ret = 0; @@ -64,7 +63,6 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans, return -ENOMEM; level = btrfs_header_level(root->node); - orig_level = level; if (level == 0) goto out; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index fb102a9..a29f193 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -786,7 +786,6 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, { struct inode *dir; int ret; - struct btrfs_key location; struct btrfs_inode_ref *ref; struct btrfs_dir_item *di; struct inode *inode; @@ -795,10 +794,6 @@ static noinline int add_inode_ref(struct btrfs_trans_handle *trans, unsigned long ref_ptr; unsigned long ref_end; - location.objectid = key->objectid; - location.type = BTRFS_INODE_ITEM_KEY; - location.offset = 0; - /* * it is possible that we didn't log all the parent directories * for a given inode. If we don't find the dir, just don't @@ -1583,7 +1578,6 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, struct btrfs_path *path; struct btrfs_root *root = wc->replay_dest; struct btrfs_key key; - u32 item_size; int level; int i; int ret; @@ -1601,7 +1595,6 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, nritems = btrfs_header_nritems(eb); for (i = 0; i < nritems; i++) { btrfs_item_key_to_cpu(eb, &key, i); - item_size = btrfs_item_size_nr(eb, i); /* inode keys are done during the first stage */ if (key.type == BTRFS_INODE_ITEM_KEY && @@ -1668,7 +1661,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, struct walk_control *wc) { u64 root_owner; - u64 root_gen; u64 bytenr; u64 ptr_gen; struct extent_buffer *next; @@ -1698,7 +1690,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, parent = path->nodes[*level]; root_owner = btrfs_header_owner(parent); - root_gen = btrfs_header_generation(parent); next = btrfs_find_create_tree_block(root, bytenr, blocksize); @@ -1749,7 +1740,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, struct walk_control *wc) { u64 root_owner; - u64 root_gen; int i; int slot; int ret; @@ -1757,8 +1747,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, for (i = *level; i < BTRFS_MAX_LEVEL - 1 && path->nodes[i]; i++) { slot = path->slots[i]; if (slot + 1 < btrfs_header_nritems(path->nodes[i])) { - struct extent_buffer *node; - node = path->nodes[i]; path->slots[i]++; *level = i; WARN_ON(*level == 0); @@ -1771,7 +1759,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, parent = path->nodes[*level + 1]; root_owner = btrfs_header_owner(parent); - root_gen = btrfs_header_generation(parent); wc->process_func(root, path->nodes[*level], wc, btrfs_header_generation(path->nodes[*level])); if (wc->free) { @@ -2273,7 +2260,7 @@ fail: } btrfs_end_log_trans(root); - return 0; + return err; } /* see comments for btrfs_del_dir_entries_in_log */ @@ -2729,7 +2716,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, struct btrfs_key max_key; struct btrfs_root *log = root->log_root; struct extent_buffer *src = NULL; - u32 size; int err = 0; int ret; int nritems; @@ -2793,7 +2779,6 @@ again: break; src = path->nodes[0]; - size = btrfs_item_size_nr(src, path->slots[0]); if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) { ins_nr++; goto next_slot; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index e25e46a..cc04dc1 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1898,7 +1898,6 @@ int btrfs_balance(struct btrfs_root *dev_root) u64 size_to_free; struct btrfs_path *path; struct btrfs_key key; - struct btrfs_chunk *chunk; struct btrfs_root *chunk_root = dev_root->fs_info->chunk_root; struct btrfs_trans_handle *trans; struct btrfs_key found_key; @@ -1962,9 +1961,6 @@ int btrfs_balance(struct btrfs_root *dev_root) if (found_key.objectid != key.objectid) break; - chunk = btrfs_item_ptr(path->nodes[0], - path->slots[0], - struct btrfs_chunk); /* chunk zero is special */ if (found_key.offset == 0) break; @@ -3031,8 +3027,7 @@ int btrfs_map_bio(struct btrfs_root *root, int rw, struct bio *bio, } bio->bi_sector = multi->stripes[dev_nr].physical >> 9; dev = multi->stripes[dev_nr].dev; - BUG_ON(rw == WRITE && !dev->writeable); - if (dev && dev->bdev) { + if (dev && dev->bdev && (rw != WRITE || dev->writeable)) { bio->bi_bdev = dev->bdev; if (async_submit) schedule_bio(root, dev, rw, bio); diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index 88ecbb2..698fdd2 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -178,7 +178,6 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) struct inode *inode = dentry->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_path *path; - struct btrfs_item *item; struct extent_buffer *leaf; struct btrfs_dir_item *di; int ret = 0, slot, advance; @@ -234,7 +233,6 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size) } advance = 1; - item = btrfs_item_nr(leaf, slot); btrfs_item_key_to_cpu(leaf, &found_key, slot); /* check to make sure this item is what we want */ diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index 3e2b90e..b9cd544 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -199,8 +199,6 @@ int btrfs_zlib_compress_pages(struct address_space *mapping, int nr_pages = 0; struct page *in_page = NULL; struct page *out_page = NULL; - int out_written = 0; - int in_read = 0; unsigned long bytes_left; *out_pages = 0; @@ -233,9 +231,6 @@ int btrfs_zlib_compress_pages(struct address_space *mapping, workspace->def_strm.avail_out = PAGE_CACHE_SIZE; workspace->def_strm.avail_in = min(len, PAGE_CACHE_SIZE); - out_written = 0; - in_read = 0; - while (workspace->def_strm.total_in < len) { ret = zlib_deflate(&workspace->def_strm, Z_SYNC_FLUSH); if (ret != Z_OK) { diff --git a/fs/ceph/super.c b/fs/ceph/super.c index d6e0e04..08b460a 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -635,7 +635,7 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc, /* * mount: join the ceph cluster, and open root directory. */ -static int ceph_mount(struct ceph_fs_client *fsc, struct vfsmount *mnt, +static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc, const char *path) { int err; @@ -678,16 +678,14 @@ static int ceph_mount(struct ceph_fs_client *fsc, struct vfsmount *mnt, } } - mnt->mnt_root = root; - mnt->mnt_sb = fsc->sb; - fsc->mount_state = CEPH_MOUNT_MOUNTED; dout("mount success\n"); - err = 0; + mutex_unlock(&fsc->client->mount_mutex); + return root; out: mutex_unlock(&fsc->client->mount_mutex); - return err; + return ERR_PTR(err); fail: if (first) { @@ -777,41 +775,45 @@ static int ceph_register_bdi(struct super_block *sb, return err; } -static int ceph_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *ceph_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { struct super_block *sb; struct ceph_fs_client *fsc; + struct dentry *res; int err; int (*compare_super)(struct super_block *, void *) = ceph_compare_super; const char *path = NULL; struct ceph_mount_options *fsopt = NULL; struct ceph_options *opt = NULL; - dout("ceph_get_sb\n"); + dout("ceph_mount\n"); err = parse_mount_options(&fsopt, &opt, flags, data, dev_name, &path); - if (err < 0) + if (err < 0) { + res = ERR_PTR(err); goto out_final; + } /* create client (which we may/may not use) */ fsc = create_fs_client(fsopt, opt); if (IS_ERR(fsc)) { - err = PTR_ERR(fsc); + res = ERR_CAST(fsc); kfree(fsopt); kfree(opt); goto out_final; } err = ceph_mdsc_init(fsc); - if (err < 0) + if (err < 0) { + res = ERR_PTR(err); goto out; + } if (ceph_test_opt(fsc->client, NOSHARE)) compare_super = NULL; sb = sget(fs_type, compare_super, ceph_set_super, fsc); if (IS_ERR(sb)) { - err = PTR_ERR(sb); + res = ERR_CAST(sb); goto out; } @@ -823,16 +825,18 @@ static int ceph_get_sb(struct file_system_type *fs_type, } else { dout("get_sb using new client %p\n", fsc); err = ceph_register_bdi(sb, fsc); - if (err < 0) + if (err < 0) { + res = ERR_PTR(err); goto out_splat; + } } - err = ceph_mount(fsc, mnt, path); - if (err < 0) + res = ceph_real_mount(fsc, path); + if (IS_ERR(res)) goto out_splat; - dout("root %p inode %p ino %llx.%llx\n", mnt->mnt_root, - mnt->mnt_root->d_inode, ceph_vinop(mnt->mnt_root->d_inode)); - return 0; + dout("root %p inode %p ino %llx.%llx\n", res, + res->d_inode, ceph_vinop(res->d_inode)); + return res; out_splat: ceph_mdsc_close_sessions(fsc->mdsc); @@ -843,8 +847,8 @@ out: ceph_mdsc_destroy(fsc); destroy_fs_client(fsc); out_final: - dout("ceph_get_sb fail %d\n", err); - return err; + dout("ceph_mount fail %ld\n", PTR_ERR(res)); + return res; } static void ceph_kill_sb(struct super_block *s) @@ -860,7 +864,7 @@ static void ceph_kill_sb(struct super_block *s) static struct file_system_type ceph_fs_type = { .owner = THIS_MODULE, .name = "ceph", - .get_sb = ceph_get_sb, + .mount = ceph_mount, .kill_sb = ceph_kill_sb, .fs_flags = FS_RENAME_DOES_D_MOVE, }; diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 917b7d4..0ed2139 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -2,6 +2,9 @@ config CIFS tristate "CIFS support (advanced network filesystem, SMBFS successor)" depends on INET select NLS + select CRYPTO + select CRYPTO_MD5 + select CRYPTO_ARC4 help This is the client VFS module for the Common Internet File System (CIFS) protocol which is the successor to the Server Message Block diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 7ac0056..f856732 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -43,18 +43,32 @@ extern void SMBencrypt(unsigned char *passwd, const unsigned char *c8, unsigned char *p24); static int cifs_calculate_signature(const struct smb_hdr *cifs_pdu, - const struct session_key *key, char *signature) + struct TCP_Server_Info *server, char *signature) { - struct MD5Context context; + int rc; - if ((cifs_pdu == NULL) || (signature == NULL) || (key == NULL)) + if (cifs_pdu == NULL || signature == NULL || server == NULL) return -EINVAL; - cifs_MD5_init(&context); - cifs_MD5_update(&context, (char *)&key->data, key->len); - cifs_MD5_update(&context, cifs_pdu->Protocol, cifs_pdu->smb_buf_length); + if (!server->secmech.sdescmd5) { + cERROR(1, "%s: Can't generate signature\n", __func__); + return -1; + } + + rc = crypto_shash_init(&server->secmech.sdescmd5->shash); + if (rc) { + cERROR(1, "%s: Oould not init md5\n", __func__); + return rc; + } + + crypto_shash_update(&server->secmech.sdescmd5->shash, + server->session_key.response, server->session_key.len); + + crypto_shash_update(&server->secmech.sdescmd5->shash, + cifs_pdu->Protocol, cifs_pdu->smb_buf_length); + + rc = crypto_shash_final(&server->secmech.sdescmd5->shash, signature); - cifs_MD5_final(signature, &context); return 0; } @@ -79,8 +93,7 @@ int cifs_sign_smb(struct smb_hdr *cifs_pdu, struct TCP_Server_Info *server, server->sequence_number++; spin_unlock(&GlobalMid_Lock); - rc = cifs_calculate_signature(cifs_pdu, &server->session_key, - smb_signature); + rc = cifs_calculate_signature(cifs_pdu, server, smb_signature); if (rc) memset(cifs_pdu->Signature.SecuritySignature, 0, 8); else @@ -90,16 +103,28 @@ int cifs_sign_smb(struct smb_hdr *cifs_pdu, struct TCP_Server_Info *server, } static int cifs_calc_signature2(const struct kvec *iov, int n_vec, - const struct session_key *key, char *signature) + struct TCP_Server_Info *server, char *signature) { - struct MD5Context context; int i; + int rc; - if ((iov == NULL) || (signature == NULL) || (key == NULL)) + if (iov == NULL || signature == NULL || server == NULL) return -EINVAL; - cifs_MD5_init(&context); - cifs_MD5_update(&context, (char *)&key->data, key->len); + if (!server->secmech.sdescmd5) { + cERROR(1, "%s: Can't generate signature\n", __func__); + return -1; + } + + rc = crypto_shash_init(&server->secmech.sdescmd5->shash); + if (rc) { + cERROR(1, "%s: Oould not init md5\n", __func__); + return rc; + } + + crypto_shash_update(&server->secmech.sdescmd5->shash, + server->session_key.response, server->session_key.len); + for (i = 0; i < n_vec; i++) { if (iov[i].iov_len == 0) continue; @@ -112,18 +137,18 @@ static int cifs_calc_signature2(const struct kvec *iov, int n_vec, if (i == 0) { if (iov[0].iov_len <= 8) /* cmd field at offset 9 */ break; /* nothing to sign or corrupt header */ - cifs_MD5_update(&context, iov[0].iov_base+4, - iov[0].iov_len-4); + crypto_shash_update(&server->secmech.sdescmd5->shash, + iov[i].iov_base + 4, iov[i].iov_len - 4); } else - cifs_MD5_update(&context, iov[i].iov_base, iov[i].iov_len); + crypto_shash_update(&server->secmech.sdescmd5->shash, + iov[i].iov_base, iov[i].iov_len); } - cifs_MD5_final(signature, &context); + rc = crypto_shash_final(&server->secmech.sdescmd5->shash, signature); - return 0; + return rc; } - int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *server, __u32 *pexpected_response_sequence_number) { @@ -146,8 +171,7 @@ int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *server, server->sequence_number++; spin_unlock(&GlobalMid_Lock); - rc = cifs_calc_signature2(iov, n_vec, &server->session_key, - smb_signature); + rc = cifs_calc_signature2(iov, n_vec, server, smb_signature); if (rc) memset(cifs_pdu->Signature.SecuritySignature, 0, 8); else @@ -157,14 +181,14 @@ int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *server, } int cifs_verify_signature(struct smb_hdr *cifs_pdu, - const struct session_key *session_key, + struct TCP_Server_Info *server, __u32 expected_sequence_number) { unsigned int rc; char server_response_sig[8]; char what_we_think_sig_should_be[20]; - if (cifs_pdu == NULL || session_key == NULL) + if (cifs_pdu == NULL || server == NULL) return -EINVAL; if (cifs_pdu->Command == SMB_COM_NEGOTIATE) @@ -193,7 +217,7 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu, cpu_to_le32(expected_sequence_number); cifs_pdu->Signature.Sequence.Reserved = 0; - rc = cifs_calculate_signature(cifs_pdu, session_key, + rc = cifs_calculate_signature(cifs_pdu, server, what_we_think_sig_should_be); if (rc) @@ -209,18 +233,28 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu, } -/* We fill in key by putting in 40 byte array which was allocated by caller */ -int cifs_calculate_session_key(struct session_key *key, const char *rn, - const char *password) +/* first calculate 24 bytes ntlm response and then 16 byte session key */ +int setup_ntlm_response(struct cifsSesInfo *ses) { - char temp_key[16]; - if ((key == NULL) || (rn == NULL)) + unsigned int temp_len = CIFS_SESS_KEY_SIZE + CIFS_AUTH_RESP_SIZE; + char temp_key[CIFS_SESS_KEY_SIZE]; + + if (!ses) return -EINVAL; - E_md4hash(password, temp_key); - mdfour(key->data.ntlm, temp_key, 16); - memcpy(key->data.ntlm+16, rn, CIFS_SESS_KEY_SIZE); - key->len = 40; + ses->auth_key.response = kmalloc(temp_len, GFP_KERNEL); + if (!ses->auth_key.response) { + cERROR(1, "NTLM can't allocate (%u bytes) memory", temp_len); + return -ENOMEM; + } + ses->auth_key.len = temp_len; + + SMBNTencrypt(ses->password, ses->server->cryptkey, + ses->auth_key.response + CIFS_SESS_KEY_SIZE); + + E_md4hash(ses->password, temp_key); + mdfour(ses->auth_key.response, temp_key, CIFS_SESS_KEY_SIZE); + return 0; } @@ -294,15 +328,15 @@ build_avpair_blob(struct cifsSesInfo *ses, const struct nls_table *nls_cp) * two times the unicode length of a server name + * size of a timestamp (which is 8 bytes). */ - ses->tilen = size + 2 * (2 * dlen) + 2 * (2 * wlen) + 8; - ses->tiblob = kzalloc(ses->tilen, GFP_KERNEL); - if (!ses->tiblob) { - ses->tilen = 0; + ses->auth_key.len = size + 2 * (2 * dlen) + 2 * (2 * wlen) + 8; + ses->auth_key.response = kzalloc(ses->auth_key.len, GFP_KERNEL); + if (!ses->auth_key.response) { + ses->auth_key.len = 0; cERROR(1, "Challenge target info allocation failure"); return -ENOMEM; } - blobptr = ses->tiblob; + blobptr = ses->auth_key.response; attrptr = (struct ntlmssp2_name *) blobptr; attrptr->type = cpu_to_le16(NTLMSSP_AV_NB_DOMAIN_NAME); @@ -357,7 +391,7 @@ build_avpair_blob(struct cifsSesInfo *ses, const struct nls_table *nls_cp) * about target string i.e. for some, just user name might suffice. */ static int -find_domain_name(struct cifsSesInfo *ses) +find_domain_name(struct cifsSesInfo *ses, const struct nls_table *nls_cp) { unsigned int attrsize; unsigned int type; @@ -366,11 +400,11 @@ find_domain_name(struct cifsSesInfo *ses) unsigned char *blobend; struct ntlmssp2_name *attrptr; - if (!ses->tilen || !ses->tiblob) + if (!ses->auth_key.len || !ses->auth_key.response) return 0; - blobptr = ses->tiblob; - blobend = ses->tiblob + ses->tilen; + blobptr = ses->auth_key.response; + blobend = blobptr + ses->auth_key.len; while (blobptr + onesize < blobend) { attrptr = (struct ntlmssp2_name *) blobptr; @@ -386,16 +420,13 @@ find_domain_name(struct cifsSesInfo *ses) if (!attrsize) break; if (!ses->domainName) { - struct nls_table *default_nls; ses->domainName = kmalloc(attrsize + 1, GFP_KERNEL); if (!ses->domainName) return -ENOMEM; - default_nls = load_nls_default(); cifs_from_ucs2(ses->domainName, (__le16 *)blobptr, attrsize, attrsize, - default_nls, false); - unload_nls(default_nls); + nls_cp, false); break; } } @@ -405,82 +436,136 @@ find_domain_name(struct cifsSesInfo *ses) return 0; } -static int calc_ntlmv2_hash(struct cifsSesInfo *ses, +static int calc_ntlmv2_hash(struct cifsSesInfo *ses, char *ntlmv2_hash, const struct nls_table *nls_cp) { int rc = 0; int len; - char nt_hash[16]; - struct HMACMD5Context *pctxt; + char nt_hash[CIFS_NTHASH_SIZE]; wchar_t *user; wchar_t *domain; + wchar_t *server; - pctxt = kmalloc(sizeof(struct HMACMD5Context), GFP_KERNEL); - - if (pctxt == NULL) - return -ENOMEM; + if (!ses->server->secmech.sdeschmacmd5) { + cERROR(1, "calc_ntlmv2_hash: can't generate ntlmv2 hash\n"); + return -1; + } /* calculate md4 hash of password */ E_md4hash(ses->password, nt_hash); - /* convert Domainname to unicode and uppercase */ - hmac_md5_init_limK_to_64(nt_hash, 16, pctxt); + crypto_shash_setkey(ses->server->secmech.hmacmd5, nt_hash, + CIFS_NTHASH_SIZE); + + rc = crypto_shash_init(&ses->server->secmech.sdeschmacmd5->shash); + if (rc) { + cERROR(1, "calc_ntlmv2_hash: could not init hmacmd5\n"); + return rc; + } /* convert ses->userName to unicode and uppercase */ len = strlen(ses->userName); user = kmalloc(2 + (len * 2), GFP_KERNEL); - if (user == NULL) + if (user == NULL) { + cERROR(1, "calc_ntlmv2_hash: user mem alloc failure\n"); + rc = -ENOMEM; goto calc_exit_2; + } len = cifs_strtoUCS((__le16 *)user, ses->userName, len, nls_cp); UniStrupr(user); - hmac_md5_update((char *)user, 2*len, pctxt); + + crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, + (char *)user, 2 * len); /* convert ses->domainName to unicode and uppercase */ if (ses->domainName) { len = strlen(ses->domainName); domain = kmalloc(2 + (len * 2), GFP_KERNEL); - if (domain == NULL) + if (domain == NULL) { + cERROR(1, "calc_ntlmv2_hash: domain mem alloc failure"); + rc = -ENOMEM; goto calc_exit_1; + } len = cifs_strtoUCS((__le16 *)domain, ses->domainName, len, nls_cp); - /* the following line was removed since it didn't work well - with lower cased domain name that passed as an option. - Maybe converting the domain name earlier makes sense */ - /* UniStrupr(domain); */ - - hmac_md5_update((char *)domain, 2*len, pctxt); - + crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, + (char *)domain, 2 * len); kfree(domain); + } else if (ses->serverName) { + len = strlen(ses->serverName); + + server = kmalloc(2 + (len * 2), GFP_KERNEL); + if (server == NULL) { + cERROR(1, "calc_ntlmv2_hash: server mem alloc failure"); + rc = -ENOMEM; + goto calc_exit_1; + } + len = cifs_strtoUCS((__le16 *)server, ses->serverName, len, + nls_cp); + crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, + (char *)server, 2 * len); + kfree(server); } + + rc = crypto_shash_final(&ses->server->secmech.sdeschmacmd5->shash, + ntlmv2_hash); + calc_exit_1: kfree(user); calc_exit_2: - /* BB FIXME what about bytes 24 through 40 of the signing key? - compare with the NTLM example */ - hmac_md5_final(ses->ntlmv2_hash, pctxt); + return rc; +} + +static int +CalcNTLMv2_response(const struct cifsSesInfo *ses, char *ntlmv2_hash) +{ + int rc; + unsigned int offset = CIFS_SESS_KEY_SIZE + 8; + + if (!ses->server->secmech.sdeschmacmd5) { + cERROR(1, "calc_ntlmv2_hash: can't generate ntlmv2 hash\n"); + return -1; + } + + crypto_shash_setkey(ses->server->secmech.hmacmd5, + ntlmv2_hash, CIFS_HMAC_MD5_HASH_SIZE); + + rc = crypto_shash_init(&ses->server->secmech.sdeschmacmd5->shash); + if (rc) { + cERROR(1, "CalcNTLMv2_response: could not init hmacmd5"); + return rc; + } + + if (ses->server->secType == RawNTLMSSP) + memcpy(ses->auth_key.response + offset, + ses->ntlmssp->cryptkey, CIFS_SERVER_CHALLENGE_SIZE); + else + memcpy(ses->auth_key.response + offset, + ses->server->cryptkey, CIFS_SERVER_CHALLENGE_SIZE); + crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, + ses->auth_key.response + offset, ses->auth_key.len - offset); + + rc = crypto_shash_final(&ses->server->secmech.sdeschmacmd5->shash, + ses->auth_key.response + CIFS_SESS_KEY_SIZE); - kfree(pctxt); return rc; } + int -setup_ntlmv2_rsp(struct cifsSesInfo *ses, char *resp_buf, - const struct nls_table *nls_cp) +setup_ntlmv2_rsp(struct cifsSesInfo *ses, const struct nls_table *nls_cp) { int rc; - struct ntlmv2_resp *buf = (struct ntlmv2_resp *)resp_buf; - struct HMACMD5Context context; - - buf->blob_signature = cpu_to_le32(0x00000101); - buf->reserved = 0; - buf->time = cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME)); - get_random_bytes(&buf->client_chal, sizeof(buf->client_chal)); - buf->reserved2 = 0; + int baselen; + unsigned int tilen; + struct ntlmv2_resp *buf; + char ntlmv2_hash[16]; + unsigned char *tiblob = NULL; /* target info blob */ if (ses->server->secType == RawNTLMSSP) { if (!ses->domainName) { - rc = find_domain_name(ses); + rc = find_domain_name(ses, nls_cp); if (rc) { cERROR(1, "error %d finding domain name", rc); goto setup_ntlmv2_rsp_ret; @@ -490,51 +575,179 @@ setup_ntlmv2_rsp(struct cifsSesInfo *ses, char *resp_buf, rc = build_avpair_blob(ses, nls_cp); if (rc) { cERROR(1, "error %d building av pair blob", rc); - return rc; + goto setup_ntlmv2_rsp_ret; } } - /* calculate buf->ntlmv2_hash */ - rc = calc_ntlmv2_hash(ses, nls_cp); + baselen = CIFS_SESS_KEY_SIZE + sizeof(struct ntlmv2_resp); + tilen = ses->auth_key.len; + tiblob = ses->auth_key.response; + + ses->auth_key.response = kmalloc(baselen + tilen, GFP_KERNEL); + if (!ses->auth_key.response) { + rc = ENOMEM; + ses->auth_key.len = 0; + cERROR(1, "%s: Can't allocate auth blob", __func__); + goto setup_ntlmv2_rsp_ret; + } + ses->auth_key.len += baselen; + + buf = (struct ntlmv2_resp *) + (ses->auth_key.response + CIFS_SESS_KEY_SIZE); + buf->blob_signature = cpu_to_le32(0x00000101); + buf->reserved = 0; + buf->time = cpu_to_le64(cifs_UnixTimeToNT(CURRENT_TIME)); + get_random_bytes(&buf->client_chal, sizeof(buf->client_chal)); + buf->reserved2 = 0; + + memcpy(ses->auth_key.response + baselen, tiblob, tilen); + + /* calculate ntlmv2_hash */ + rc = calc_ntlmv2_hash(ses, ntlmv2_hash, nls_cp); if (rc) { cERROR(1, "could not get v2 hash rc %d", rc); goto setup_ntlmv2_rsp_ret; } - CalcNTLMv2_response(ses, resp_buf); + + /* calculate first part of the client response (CR1) */ + rc = CalcNTLMv2_response(ses, ntlmv2_hash); + if (rc) { + cERROR(1, "Could not calculate CR1 rc: %d", rc); + goto setup_ntlmv2_rsp_ret; + } /* now calculate the session key for NTLMv2 */ - hmac_md5_init_limK_to_64(ses->ntlmv2_hash, 16, &context); - hmac_md5_update(resp_buf, 16, &context); - hmac_md5_final(ses->auth_key.data.ntlmv2.key, &context); + crypto_shash_setkey(ses->server->secmech.hmacmd5, + ntlmv2_hash, CIFS_HMAC_MD5_HASH_SIZE); + + rc = crypto_shash_init(&ses->server->secmech.sdeschmacmd5->shash); + if (rc) { + cERROR(1, "%s: Could not init hmacmd5\n", __func__); + goto setup_ntlmv2_rsp_ret; + } - memcpy(&ses->auth_key.data.ntlmv2.resp, resp_buf, - sizeof(struct ntlmv2_resp)); - ses->auth_key.len = 16 + sizeof(struct ntlmv2_resp); + crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, + ses->auth_key.response + CIFS_SESS_KEY_SIZE, + CIFS_HMAC_MD5_HASH_SIZE); - return 0; + rc = crypto_shash_final(&ses->server->secmech.sdeschmacmd5->shash, + ses->auth_key.response); setup_ntlmv2_rsp_ret: - kfree(ses->tiblob); - ses->tiblob = NULL; - ses->tilen = 0; + kfree(tiblob); return rc; } -void CalcNTLMv2_response(const struct cifsSesInfo *ses, - char *v2_session_response) +int +calc_seckey(struct cifsSesInfo *ses) { - struct HMACMD5Context context; - /* rest of v2 struct already generated */ - memcpy(v2_session_response + 8, ses->cryptKey, 8); - hmac_md5_init_limK_to_64(ses->ntlmv2_hash, 16, &context); + int rc; + struct crypto_blkcipher *tfm_arc4; + struct scatterlist sgin, sgout; + struct blkcipher_desc desc; + unsigned char sec_key[CIFS_SESS_KEY_SIZE]; /* a nonce */ + + get_random_bytes(sec_key, CIFS_SESS_KEY_SIZE); + + tfm_arc4 = crypto_alloc_blkcipher("ecb(arc4)", 0, CRYPTO_ALG_ASYNC); + if (!tfm_arc4 || IS_ERR(tfm_arc4)) { + cERROR(1, "could not allocate crypto API arc4\n"); + return PTR_ERR(tfm_arc4); + } - hmac_md5_update(v2_session_response+8, - sizeof(struct ntlmv2_resp) - 8, &context); + desc.tfm = tfm_arc4; - if (ses->tilen) - hmac_md5_update(ses->tiblob, ses->tilen, &context); + crypto_blkcipher_setkey(tfm_arc4, ses->auth_key.response, + CIFS_SESS_KEY_SIZE); - hmac_md5_final(v2_session_response, &context); -/* cifs_dump_mem("v2_sess_rsp: ", v2_session_response, 32); */ + sg_init_one(&sgin, sec_key, CIFS_SESS_KEY_SIZE); + sg_init_one(&sgout, ses->ntlmssp->ciphertext, CIFS_CPHTXT_SIZE); + + rc = crypto_blkcipher_encrypt(&desc, &sgout, &sgin, CIFS_CPHTXT_SIZE); + if (rc) { + cERROR(1, "could not encrypt session key rc: %d\n", rc); + crypto_free_blkcipher(tfm_arc4); + return rc; + } + + /* make secondary_key/nonce as session key */ + memcpy(ses->auth_key.response, sec_key, CIFS_SESS_KEY_SIZE); + /* and make len as that of session key only */ + ses->auth_key.len = CIFS_SESS_KEY_SIZE; + + crypto_free_blkcipher(tfm_arc4); + + return 0; +} + +void +cifs_crypto_shash_release(struct TCP_Server_Info *server) +{ + if (server->secmech.md5) + crypto_free_shash(server->secmech.md5); + + if (server->secmech.hmacmd5) + crypto_free_shash(server->secmech.hmacmd5); + + kfree(server->secmech.sdeschmacmd5); + + kfree(server->secmech.sdescmd5); +} + +int +cifs_crypto_shash_allocate(struct TCP_Server_Info *server) +{ + int rc; + unsigned int size; + + server->secmech.hmacmd5 = crypto_alloc_shash("hmac(md5)", 0, 0); + if (!server->secmech.hmacmd5 || + IS_ERR(server->secmech.hmacmd5)) { + cERROR(1, "could not allocate crypto hmacmd5\n"); + return PTR_ERR(server->secmech.hmacmd5); + } + + server->secmech.md5 = crypto_alloc_shash("md5", 0, 0); + if (!server->secmech.md5 || IS_ERR(server->secmech.md5)) { + cERROR(1, "could not allocate crypto md5\n"); + rc = PTR_ERR(server->secmech.md5); + goto crypto_allocate_md5_fail; + } + + size = sizeof(struct shash_desc) + + crypto_shash_descsize(server->secmech.hmacmd5); + server->secmech.sdeschmacmd5 = kmalloc(size, GFP_KERNEL); + if (!server->secmech.sdeschmacmd5) { + cERROR(1, "cifs_crypto_shash_allocate: can't alloc hmacmd5\n"); + rc = -ENOMEM; + goto crypto_allocate_hmacmd5_sdesc_fail; + } + server->secmech.sdeschmacmd5->shash.tfm = server->secmech.hmacmd5; + server->secmech.sdeschmacmd5->shash.flags = 0x0; + + + size = sizeof(struct shash_desc) + + crypto_shash_descsize(server->secmech.md5); + server->secmech.sdescmd5 = kmalloc(size, GFP_KERNEL); + if (!server->secmech.sdescmd5) { + cERROR(1, "cifs_crypto_shash_allocate: can't alloc md5\n"); + rc = -ENOMEM; + goto crypto_allocate_md5_sdesc_fail; + } + server->secmech.sdescmd5->shash.tfm = server->secmech.md5; + server->secmech.sdescmd5->shash.flags = 0x0; + + return 0; + +crypto_allocate_md5_sdesc_fail: + kfree(server->secmech.sdeschmacmd5); + +crypto_allocate_hmacmd5_sdesc_fail: + crypto_free_shash(server->secmech.md5); + +crypto_allocate_md5_fail: + crypto_free_shash(server->secmech.hmacmd5); + + return rc; } diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 3437163..75c4eaa 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -318,7 +318,6 @@ cifs_alloc_inode(struct super_block *sb) return NULL; cifs_inode->cifsAttrs = 0x20; /* default */ cifs_inode->time = 0; - cifs_inode->write_behind_rc = 0; /* Until the file is open and we have gotten oplock info back from the server, can not assume caching of file data or metadata */ @@ -545,9 +544,9 @@ static const struct super_operations cifs_super_ops = { #endif }; -static int -cifs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry * +cifs_do_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { int rc; struct super_block *sb; @@ -557,18 +556,17 @@ cifs_get_sb(struct file_system_type *fs_type, cFYI(1, "Devname: %s flags: %d ", dev_name, flags); if (IS_ERR(sb)) - return PTR_ERR(sb); + return ERR_CAST(sb); sb->s_flags = flags; rc = cifs_read_super(sb, data, dev_name, flags & MS_SILENT ? 1 : 0); if (rc) { deactivate_locked_super(sb); - return rc; + return ERR_PTR(rc); } sb->s_flags |= MS_ACTIVE; - simple_set_mnt(mnt, sb); - return 0; + return dget(sb->s_root); } static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, @@ -634,7 +632,7 @@ static int cifs_setlease(struct file *file, long arg, struct file_lock **lease) struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", - .get_sb = cifs_get_sb, + .mount = cifs_do_mount, .kill_sb = kill_anon_super, /* .fs_flags */ }; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index f35795a..897b2b2 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -112,5 +112,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* EXPERIMENTAL */ -#define CIFS_VERSION "1.67" +#define CIFS_VERSION "1.68" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 3365e77f..f259e4d 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -25,6 +25,9 @@ #include <linux/workqueue.h> #include "cifs_fs_sb.h" #include "cifsacl.h" +#include <crypto/internal/hash.h> +#include <linux/scatterlist.h> + /* * The sizes of various internal tables and strings */ @@ -74,7 +77,7 @@ * CIFS vfs client Status information (based on what we know.) */ - /* associated with each tcp and smb session */ +/* associated with each tcp and smb session */ enum statusEnum { CifsNew = 0, CifsGood, @@ -99,14 +102,29 @@ enum protocolEnum { struct session_key { unsigned int len; - union { - char ntlm[CIFS_SESS_KEY_SIZE + 16]; - char krb5[CIFS_SESS_KEY_SIZE + 16]; /* BB: length correct? */ - struct { - char key[16]; - struct ntlmv2_resp resp; - } ntlmv2; - } data; + char *response; +}; + +/* crypto security descriptor definition */ +struct sdesc { + struct shash_desc shash; + char ctx[]; +}; + +/* crypto hashing related structure/fields, not specific to a sec mech */ +struct cifs_secmech { + struct crypto_shash *hmacmd5; /* hmac-md5 hash function */ + struct crypto_shash *md5; /* md5 hash function */ + struct sdesc *sdeschmacmd5; /* ctxt to generate ntlmv2 hash, CR1 */ + struct sdesc *sdescmd5; /* ctxt to generate cifs/smb signature */ +}; + +/* per smb session structure/fields */ +struct ntlmssp_auth { + __u32 client_flags; /* sent by client in type 1 ntlmsssp exchange */ + __u32 server_flags; /* sent by server in type 2 ntlmssp exchange */ + unsigned char ciphertext[CIFS_CPHTXT_SIZE]; /* sent to server */ + char cryptkey[CIFS_CRYPTO_KEY_SIZE]; /* used by ntlmssp */ }; struct cifs_cred { @@ -179,12 +197,14 @@ struct TCP_Server_Info { int capabilities; /* allow selective disabling of caps by smb sess */ int timeAdj; /* Adjust for difference in server time zone in sec */ __u16 CurrentMid; /* multiplex id - rotating counter */ + char cryptkey[CIFS_CRYPTO_KEY_SIZE]; /* used by ntlm, ntlmv2 etc */ /* 16th byte of RFC1001 workstation name is always null */ char workstation_RFC1001_name[RFC1001_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* needed for CIFS PDU signature */ struct session_key session_key; unsigned long lstrp; /* when we got last response from this server */ u16 dialect; /* dialect index that server chose */ + struct cifs_secmech secmech; /* crypto sec mech functs, descriptors */ /* extended security flavors that server supports */ bool sec_kerberos; /* supports plain Kerberos */ bool sec_mskerberos; /* supports legacy MS Kerberos */ @@ -222,11 +242,8 @@ struct cifsSesInfo { char userName[MAX_USERNAME_SIZE + 1]; char *domainName; char *password; - char cryptKey[CIFS_CRYPTO_KEY_SIZE]; struct session_key auth_key; - char ntlmv2_hash[16]; - unsigned int tilen; /* length of the target info blob */ - unsigned char *tiblob; /* target info blob in challenge response */ + struct ntlmssp_auth *ntlmssp; /* ciphertext, flags, server challenge */ bool need_reconnect:1; /* connection reset, uid now invalid */ }; /* no more than one of the following three session flags may be set */ @@ -395,16 +412,19 @@ struct cifsFileInfo { struct list_head llist; /* list of byte range locks we have. */ bool invalidHandle:1; /* file closed via session abend */ bool oplock_break_cancelled:1; - atomic_t count; /* reference count */ + int count; /* refcount protected by cifs_file_list_lock */ struct mutex fh_mutex; /* prevents reopen race after dead ses*/ struct cifs_search_info srch_inf; struct work_struct oplock_break; /* work for oplock breaks */ }; -/* Take a reference on the file private data */ +/* + * Take a reference on the file private data. Must be called with + * cifs_file_list_lock held. + */ static inline void cifsFileInfo_get(struct cifsFileInfo *cifs_file) { - atomic_inc(&cifs_file->count); + ++cifs_file->count; } void cifsFileInfo_put(struct cifsFileInfo *cifs_file); @@ -417,7 +437,6 @@ struct cifsInodeInfo { struct list_head lockList; /* BB add in lists for dirty pages i.e. write caching info for oplock */ struct list_head openFileList; - int write_behind_rc; __u32 cifsAttrs; /* e.g. DOS archive bit, sparse, compressed, system */ unsigned long time; /* jiffies of last update/check of inode */ bool clientCanCacheRead:1; /* read oplock */ @@ -668,7 +687,7 @@ require use of the stronger protocol */ * GlobalMid_Lock protects: * list operations on pending_mid_q and oplockQ * updates to XID counters, multiplex id and SMB sequence numbers - * GlobalSMBSesLock protects: + * cifs_file_list_lock protects: * list operations on tcp and SMB session lists and tCon lists * f_owner.lock protects certain per file struct operations * mapping->page_lock protects certain per page operations diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index b0f4b56..de36b09 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -131,9 +131,20 @@ #define CIFS_CRYPTO_KEY_SIZE (8) /* + * Size of the ntlm client response + */ +#define CIFS_AUTH_RESP_SIZE (24) + +/* * Size of the session key (crypto key encrypted with the password */ -#define CIFS_SESS_KEY_SIZE (24) +#define CIFS_SESS_KEY_SIZE (16) + +#define CIFS_CLIENT_CHALLENGE_SIZE (8) +#define CIFS_SERVER_CHALLENGE_SIZE (8) +#define CIFS_HMAC_MD5_HASH_SIZE (16) +#define CIFS_CPHTXT_SIZE (16) +#define CIFS_NTHASH_SIZE (16) /* * Maximum user name length diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index e593c40..edb6d90 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -362,13 +362,15 @@ extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *, __u32 *); extern int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *, __u32 *); extern int cifs_verify_signature(struct smb_hdr *, - const struct session_key *session_key, + struct TCP_Server_Info *server, __u32 expected_sequence_number); -extern int cifs_calculate_session_key(struct session_key *key, const char *rn, - const char *pass); -extern void CalcNTLMv2_response(const struct cifsSesInfo *, char *); -extern int setup_ntlmv2_rsp(struct cifsSesInfo *, char *, - const struct nls_table *); +extern void SMBNTencrypt(unsigned char *, unsigned char *, unsigned char *); +extern int setup_ntlm_response(struct cifsSesInfo *); +extern int setup_ntlmv2_rsp(struct cifsSesInfo *, const struct nls_table *); +extern int cifs_crypto_shash_allocate(struct TCP_Server_Info *); +extern void cifs_crypto_shash_release(struct TCP_Server_Info *); +extern int calc_seckey(struct cifsSesInfo *); + #ifdef CONFIG_CIFS_WEAK_PW_HASH extern void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt, char *lnm_session_key); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index e98f1f3..2f2632b 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -503,7 +503,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) if (rsp->EncryptionKeyLength == cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { - memcpy(ses->cryptKey, rsp->EncryptionKey, + memcpy(ses->server->cryptkey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if (server->secMode & SECMODE_PW_ENCRYPT) { rc = -EIO; /* need cryptkey unless plain text */ @@ -574,7 +574,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); server->timeAdj *= 60; if (pSMBr->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { - memcpy(ses->cryptKey, pSMBr->u.EncryptionKey, + memcpy(ses->server->cryptkey, pSMBr->u.EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if ((pSMBr->hdr.Flags2 & SMBFLG2_EXT_SEC) && (pSMBr->EncryptionKeyLength == 0)) { diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 7e73176..9eb327d 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -175,6 +175,9 @@ cifs_reconnect(struct TCP_Server_Info *server) } server->sequence_number = 0; server->session_estab = false; + kfree(server->session_key.response); + server->session_key.response = NULL; + server->session_key.len = 0; spin_lock(&GlobalMid_Lock); list_for_each(tmp, &server->pending_mid_q) { @@ -1064,7 +1067,7 @@ cifs_parse_mount_options(char *options, const char *devname, } i = cifs_convert_address((struct sockaddr *)&vol->srcaddr, value, strlen(value)); - if (i < 0) { + if (i == 0) { printk(KERN_WARNING "CIFS: Could not parse" " srcaddr: %s\n", value); @@ -1560,8 +1563,13 @@ cifs_put_tcp_session(struct TCP_Server_Info *server) server->tcpStatus = CifsExiting; spin_unlock(&GlobalMid_Lock); + cifs_crypto_shash_release(server); cifs_fscache_release_client_cookie(server); + kfree(server->session_key.response); + server->session_key.response = NULL; + server->session_key.len = 0; + task = xchg(&server->tsk, NULL); if (task) force_sig(SIGKILL, task); @@ -1614,10 +1622,16 @@ cifs_get_tcp_session(struct smb_vol *volume_info) goto out_err; } + rc = cifs_crypto_shash_allocate(tcp_ses); + if (rc) { + cERROR(1, "could not setup hash structures rc %d", rc); + goto out_err; + } + tcp_ses->hostname = extract_hostname(volume_info->UNC); if (IS_ERR(tcp_ses->hostname)) { rc = PTR_ERR(tcp_ses->hostname); - goto out_err; + goto out_err_crypto_release; } tcp_ses->noblocksnd = volume_info->noblocksnd; @@ -1661,7 +1675,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info) } if (rc < 0) { cERROR(1, "Error connecting to socket. Aborting operation"); - goto out_err; + goto out_err_crypto_release; } /* @@ -1675,7 +1689,7 @@ cifs_get_tcp_session(struct smb_vol *volume_info) rc = PTR_ERR(tcp_ses->tsk); cERROR(1, "error %d create cifsd thread", rc); module_put(THIS_MODULE); - goto out_err; + goto out_err_crypto_release; } /* thread spawned, put it on the list */ @@ -1687,6 +1701,9 @@ cifs_get_tcp_session(struct smb_vol *volume_info) return tcp_ses; +out_err_crypto_release: + cifs_crypto_shash_release(tcp_ses); + out_err: if (tcp_ses) { if (!IS_ERR(tcp_ses->hostname)) @@ -1801,8 +1818,6 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) if (ses == NULL) goto get_ses_fail; - ses->tilen = 0; - ses->tiblob = NULL; /* new SMB session uses our server ref */ ses->server = server; if (server->addr.sockAddr6.sin6_family == AF_INET6) @@ -1823,10 +1838,9 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) goto get_ses_fail; } if (volume_info->domainname) { - int len = strlen(volume_info->domainname); - ses->domainName = kmalloc(len + 1, GFP_KERNEL); - if (ses->domainName) - strcpy(ses->domainName, volume_info->domainname); + ses->domainName = kstrdup(volume_info->domainname, GFP_KERNEL); + if (!ses->domainName) + goto get_ses_fail; } ses->cred_uid = volume_info->cred_uid; ses->linux_uid = volume_info->linux_uid; @@ -2985,13 +2999,13 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, #ifdef CONFIG_CIFS_WEAK_PW_HASH if ((global_secflags & CIFSSEC_MAY_LANMAN) && (ses->server->secType == LANMAN)) - calc_lanman_hash(tcon->password, ses->cryptKey, + calc_lanman_hash(tcon->password, ses->server->cryptkey, ses->server->secMode & SECMODE_PW_ENCRYPT ? true : false, bcc_ptr); else #endif /* CIFS_WEAK_PW_HASH */ - SMBNTencrypt(tcon->password, ses->cryptKey, bcc_ptr); + SMBNTencrypt(tcon->password, ses->server->cryptkey, bcc_ptr); bcc_ptr += CIFS_SESS_KEY_SIZE; if (ses->capabilities & CAP_UNICODE) { @@ -3178,10 +3192,11 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, } else { mutex_lock(&ses->server->srv_mutex); if (!server->session_estab) { - memcpy(&server->session_key.data, - &ses->auth_key.data, ses->auth_key.len); + server->session_key.response = ses->auth_key.response; server->session_key.len = ses->auth_key.len; - ses->server->session_estab = true; + server->sequence_number = 0x2; + server->session_estab = true; + ses->auth_key.response = NULL; } mutex_unlock(&server->srv_mutex); @@ -3192,6 +3207,12 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *ses, spin_unlock(&GlobalMid_Lock); } + kfree(ses->auth_key.response); + ses->auth_key.response = NULL; + ses->auth_key.len = 0; + kfree(ses->ntlmssp); + ses->ntlmssp = NULL; + return rc; } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 45af003..ae82159 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -131,8 +131,7 @@ static inline int cifs_open_inode_helper(struct inode *inode, /* BB no need to lock inode until after invalidate since namei code should already have it locked? */ rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) - pCifsInode->write_behind_rc = rc; + mapping_set_error(inode->i_mapping, rc); } cFYI(1, "invalidating remote inode since open detected it " "changed"); @@ -232,6 +231,7 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, if (pCifsFile == NULL) return pCifsFile; + pCifsFile->count = 1; pCifsFile->netfid = fileHandle; pCifsFile->pid = current->tgid; pCifsFile->uid = current_fsuid(); @@ -242,7 +242,6 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, mutex_init(&pCifsFile->fh_mutex); mutex_init(&pCifsFile->lock_mutex); INIT_LIST_HEAD(&pCifsFile->llist); - atomic_set(&pCifsFile->count, 1); INIT_WORK(&pCifsFile->oplock_break, cifs_oplock_break); spin_lock(&cifs_file_list_lock); @@ -267,7 +266,8 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, /* * Release a reference on the file private data. This may involve closing - * the filehandle out on the server. + * the filehandle out on the server. Must be called without holding + * cifs_file_list_lock. */ void cifsFileInfo_put(struct cifsFileInfo *cifs_file) { @@ -276,7 +276,7 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file) struct cifsLockInfo *li, *tmp; spin_lock(&cifs_file_list_lock); - if (!atomic_dec_and_test(&cifs_file->count)) { + if (--cifs_file->count > 0) { spin_unlock(&cifs_file_list_lock); return; } @@ -605,8 +605,7 @@ reopen_success: if (can_flush) { rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) - CIFS_I(inode)->write_behind_rc = rc; + mapping_set_error(inode->i_mapping, rc); pCifsInode->clientCanCacheAll = false; pCifsInode->clientCanCacheRead = false; @@ -1353,6 +1352,7 @@ static int cifs_writepages(struct address_space *mapping, if (!experimEnabled && tcon->ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { cifsFileInfo_put(open_file); + kfree(iov); return generic_writepages(mapping, wbc); } cifsFileInfo_put(open_file); @@ -1478,12 +1478,7 @@ retry: if (rc || bytes_written < bytes_to_write) { cERROR(1, "Write2 ret %d, wrote %d", rc, bytes_written); - /* BB what if continued retry is - requested via mount flags? */ - if (rc == -ENOSPC) - set_bit(AS_ENOSPC, &mapping->flags); - else - set_bit(AS_EIO, &mapping->flags); + mapping_set_error(mapping, rc); } else { cifs_stats_bytes_written(tcon, bytes_written); } @@ -1628,11 +1623,10 @@ int cifs_fsync(struct file *file, int datasync) rc = filemap_write_and_wait(inode->i_mapping); if (rc == 0) { - rc = CIFS_I(inode)->write_behind_rc; - CIFS_I(inode)->write_behind_rc = 0; + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); + tcon = tlink_tcon(smbfile->tlink); - if (!rc && tcon && smbfile && - !(CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) rc = CIFSSMBFlush(xid, tcon, smbfile->netfid); } @@ -1677,21 +1671,8 @@ int cifs_flush(struct file *file, fl_owner_t id) struct inode *inode = file->f_path.dentry->d_inode; int rc = 0; - /* Rather than do the steps manually: - lock the inode for writing - loop through pages looking for write behind data (dirty pages) - coalesce into contiguous 16K (or smaller) chunks to write to server - send to server (prefer in parallel) - deal with writebehind errors - unlock inode for writing - filemapfdatawrite appears easier for the time being */ - - rc = filemap_fdatawrite(inode->i_mapping); - /* reset wb rc if we were able to write out dirty pages */ - if (!rc) { - rc = CIFS_I(inode)->write_behind_rc; - CIFS_I(inode)->write_behind_rc = 0; - } + if (file->f_mode & FMODE_WRITE) + rc = filemap_write_and_wait(inode->i_mapping); cFYI(1, "Flush inode %p file %p rc %d", inode, file, rc); @@ -2270,7 +2251,7 @@ void cifs_oplock_break(struct work_struct *work) oplock_break); struct inode *inode = cfile->dentry->d_inode; struct cifsInodeInfo *cinode = CIFS_I(inode); - int rc, waitrc = 0; + int rc = 0; if (inode && S_ISREG(inode->i_mode)) { if (cinode->clientCanCacheRead) @@ -2279,13 +2260,10 @@ void cifs_oplock_break(struct work_struct *work) break_lease(inode, O_WRONLY); rc = filemap_fdatawrite(inode->i_mapping); if (cinode->clientCanCacheRead == 0) { - waitrc = filemap_fdatawait(inode->i_mapping); + rc = filemap_fdatawait(inode->i_mapping); + mapping_set_error(inode->i_mapping, rc); invalidate_remote_inode(inode); } - if (!rc) - rc = waitrc; - if (rc) - cinode->write_behind_rc = rc; cFYI(1, "Oplock flush inode %p rc %d", inode, rc); } @@ -2304,7 +2282,7 @@ void cifs_oplock_break(struct work_struct *work) /* * We might have kicked in before is_valid_oplock_break() * finished grabbing reference for us. Make sure it's done by - * waiting for GlobalSMSSeslock. + * waiting for cifs_file_list_lock. */ spin_lock(&cifs_file_list_lock); spin_unlock(&cifs_file_list_lock); @@ -2312,6 +2290,7 @@ void cifs_oplock_break(struct work_struct *work) cifs_oplock_break_put(cfile); } +/* must be called while holding cifs_file_list_lock */ void cifs_oplock_break_get(struct cifsFileInfo *cfile) { cifs_sb_active(cfile->dentry->d_sb); diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 9497930..39869c3c 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1682,8 +1682,7 @@ cifs_invalidate_mapping(struct inode *inode) /* write back any cached data */ if (inode->i_mapping && inode->i_mapping->nrpages != 0) { rc = filemap_write_and_wait(inode->i_mapping); - if (rc) - cifs_i->write_behind_rc = rc; + mapping_set_error(inode->i_mapping, rc); } invalidate_remote_inode(inode); cifs_fscache_reset_inode_cookie(inode); @@ -1943,10 +1942,8 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) * the flush returns error? */ rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) { - cifsInode->write_behind_rc = rc; - rc = 0; - } + mapping_set_error(inode->i_mapping, rc); + rc = 0; if (attrs->ia_valid & ATTR_SIZE) { rc = cifs_set_file_size(inode, attrs, xid, full_path); @@ -2087,10 +2084,8 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) * the flush returns error? */ rc = filemap_write_and_wait(inode->i_mapping); - if (rc != 0) { - cifsInode->write_behind_rc = rc; - rc = 0; - } + mapping_set_error(inode->i_mapping, rc); + rc = 0; if (attrs->ia_valid & ATTR_SIZE) { rc = cifs_set_file_size(inode, attrs, xid, full_path); diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 1c681f6..c4e296f 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -577,7 +577,7 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) * cifs_oplock_break_put() can't be called * from here. Get reference after queueing * succeeded. cifs_oplock_break() will - * synchronize using GlobalSMSSeslock. + * synchronize using cifs_file_list_lock. */ if (queue_work(system_nrt_wq, &netfile->oplock_break)) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 2a11efd..7b01d3f 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -32,9 +32,6 @@ #include <linux/slab.h> #include "cifs_spnego.h" -extern void SMBNTencrypt(unsigned char *passwd, unsigned char *c8, - unsigned char *p24); - /* * Checks if this is the first smb session to be reconnected after * the socket has been reestablished (so we know whether to use vc 0). @@ -402,23 +399,22 @@ static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, return -EINVAL; } - memcpy(ses->cryptKey, pblob->Challenge, CIFS_CRYPTO_KEY_SIZE); + memcpy(ses->ntlmssp->cryptkey, pblob->Challenge, CIFS_CRYPTO_KEY_SIZE); /* BB we could decode pblob->NegotiateFlags; some may be useful */ /* In particular we can examine sign flags */ /* BB spec says that if AvId field of MsvAvTimestamp is populated then we must set the MIC field of the AUTHENTICATE_MESSAGE */ - + ses->ntlmssp->server_flags = le32_to_cpu(pblob->NegotiateFlags); tioffset = cpu_to_le16(pblob->TargetInfoArray.BufferOffset); tilen = cpu_to_le16(pblob->TargetInfoArray.Length); - ses->tilen = tilen; - if (ses->tilen) { - ses->tiblob = kmalloc(tilen, GFP_KERNEL); - if (!ses->tiblob) { + if (tilen) { + ses->auth_key.response = kmalloc(tilen, GFP_KERNEL); + if (!ses->auth_key.response) { cERROR(1, "Challenge target info allocation failure"); - ses->tilen = 0; return -ENOMEM; } - memcpy(ses->tiblob, bcc_ptr + tioffset, ses->tilen); + memcpy(ses->auth_key.response, bcc_ptr + tioffset, tilen); + ses->auth_key.len = tilen; } return 0; @@ -443,10 +439,12 @@ static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_NEGOTIATE_NTLM; if (ses->server->secMode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { flags |= NTLMSSP_NEGOTIATE_SIGN; - if (ses->server->secMode & SECMODE_SIGN_REQUIRED) - flags |= NTLMSSP_NEGOTIATE_ALWAYS_SIGN; + if (!ses->server->session_estab) + flags |= NTLMSSP_NEGOTIATE_KEY_XCH | + NTLMSSP_NEGOTIATE_EXTENDED_SEC; + } sec_blob->NegotiateFlags |= cpu_to_le32(flags); @@ -469,11 +467,9 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, const struct nls_table *nls_cp) { int rc; - unsigned int size; AUTHENTICATE_MESSAGE *sec_blob = (AUTHENTICATE_MESSAGE *)pbuffer; __u32 flags; unsigned char *tmp; - struct ntlmv2_resp ntlmv2_response = {}; memcpy(sec_blob->Signature, NTLMSSP_SIGNATURE, 8); sec_blob->MessageType = NtLmAuthenticate; @@ -497,25 +493,19 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, sec_blob->LmChallengeResponse.MaximumLength = 0; sec_blob->NtChallengeResponse.BufferOffset = cpu_to_le32(tmp - pbuffer); - rc = setup_ntlmv2_rsp(ses, (char *)&ntlmv2_response, nls_cp); + rc = setup_ntlmv2_rsp(ses, nls_cp); if (rc) { cERROR(1, "Error %d during NTLMSSP authentication", rc); goto setup_ntlmv2_ret; } - size = sizeof(struct ntlmv2_resp); - memcpy(tmp, (char *)&ntlmv2_response, size); - tmp += size; - if (ses->tilen > 0) { - memcpy(tmp, ses->tiblob, ses->tilen); - tmp += ses->tilen; - } + memcpy(tmp, ses->auth_key.response + CIFS_SESS_KEY_SIZE, + ses->auth_key.len - CIFS_SESS_KEY_SIZE); + tmp += ses->auth_key.len - CIFS_SESS_KEY_SIZE; - sec_blob->NtChallengeResponse.Length = cpu_to_le16(size + ses->tilen); + sec_blob->NtChallengeResponse.Length = + cpu_to_le16(ses->auth_key.len - CIFS_SESS_KEY_SIZE); sec_blob->NtChallengeResponse.MaximumLength = - cpu_to_le16(size + ses->tilen); - kfree(ses->tiblob); - ses->tiblob = NULL; - ses->tilen = 0; + cpu_to_le16(ses->auth_key.len - CIFS_SESS_KEY_SIZE); if (ses->domainName == NULL) { sec_blob->DomainName.BufferOffset = cpu_to_le32(tmp - pbuffer); @@ -554,9 +544,19 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, sec_blob->WorkstationName.MaximumLength = 0; tmp += 2; - sec_blob->SessionKey.BufferOffset = cpu_to_le32(tmp - pbuffer); - sec_blob->SessionKey.Length = 0; - sec_blob->SessionKey.MaximumLength = 0; + if ((ses->ntlmssp->server_flags & NTLMSSP_NEGOTIATE_KEY_XCH) && + !calc_seckey(ses)) { + memcpy(tmp, ses->ntlmssp->ciphertext, CIFS_CPHTXT_SIZE); + sec_blob->SessionKey.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->SessionKey.Length = cpu_to_le16(CIFS_CPHTXT_SIZE); + sec_blob->SessionKey.MaximumLength = + cpu_to_le16(CIFS_CPHTXT_SIZE); + tmp += CIFS_CPHTXT_SIZE; + } else { + sec_blob->SessionKey.BufferOffset = cpu_to_le32(tmp - pbuffer); + sec_blob->SessionKey.Length = 0; + sec_blob->SessionKey.MaximumLength = 0; + } setup_ntlmv2_ret: *buflen = tmp - pbuffer; @@ -600,8 +600,16 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, return -EINVAL; type = ses->server->secType; - cFYI(1, "sess setup type %d", type); + if (type == RawNTLMSSP) { + /* if memory allocation is successful, caller of this function + * frees it. + */ + ses->ntlmssp = kmalloc(sizeof(struct ntlmssp_auth), GFP_KERNEL); + if (!ses->ntlmssp) + return -ENOMEM; + } + ssetup_ntlmssp_authenticate: if (phase == NtLmChallenge) phase = NtLmAuthenticate; /* if ntlmssp, now final phase */ @@ -666,10 +674,14 @@ ssetup_ntlmssp_authenticate: /* no capabilities flags in old lanman negotiation */ pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_SESS_KEY_SIZE); - /* BB calculate hash with password */ - /* and copy into bcc */ - calc_lanman_hash(ses->password, ses->cryptKey, + /* Calculate hash with password and copy into bcc_ptr. + * Encryption Key (stored as in cryptkey) gets used if the + * security mode bit in Negottiate Protocol response states + * to use challenge/response method (i.e. Password bit is 1). + */ + + calc_lanman_hash(ses->password, ses->server->cryptkey, ses->server->secMode & SECMODE_PW_ENCRYPT ? true : false, lnm_session_key); @@ -687,24 +699,27 @@ ssetup_ntlmssp_authenticate: ascii_ssetup_strings(&bcc_ptr, ses, nls_cp); #endif } else if (type == NTLM) { - char ntlm_session_key[CIFS_SESS_KEY_SIZE]; - pSMB->req_no_secext.Capabilities = cpu_to_le32(capabilities); pSMB->req_no_secext.CaseInsensitivePasswordLength = - cpu_to_le16(CIFS_SESS_KEY_SIZE); + cpu_to_le16(CIFS_AUTH_RESP_SIZE); pSMB->req_no_secext.CaseSensitivePasswordLength = - cpu_to_le16(CIFS_SESS_KEY_SIZE); + cpu_to_le16(CIFS_AUTH_RESP_SIZE); + + /* calculate ntlm response and session key */ + rc = setup_ntlm_response(ses); + if (rc) { + cERROR(1, "Error %d during NTLM authentication", rc); + goto ssetup_exit; + } - /* calculate session key */ - SMBNTencrypt(ses->password, ses->cryptKey, ntlm_session_key); + /* copy ntlm response */ + memcpy(bcc_ptr, ses->auth_key.response + CIFS_SESS_KEY_SIZE, + CIFS_AUTH_RESP_SIZE); + bcc_ptr += CIFS_AUTH_RESP_SIZE; + memcpy(bcc_ptr, ses->auth_key.response + CIFS_SESS_KEY_SIZE, + CIFS_AUTH_RESP_SIZE); + bcc_ptr += CIFS_AUTH_RESP_SIZE; - cifs_calculate_session_key(&ses->auth_key, - ntlm_session_key, ses->password); - /* copy session key */ - memcpy(bcc_ptr, (char *)ntlm_session_key, CIFS_SESS_KEY_SIZE); - bcc_ptr += CIFS_SESS_KEY_SIZE; - memcpy(bcc_ptr, (char *)ntlm_session_key, CIFS_SESS_KEY_SIZE); - bcc_ptr += CIFS_SESS_KEY_SIZE; if (ses->capabilities & CAP_UNICODE) { /* unicode strings must be word aligned */ if (iov[0].iov_len % 2) { @@ -715,47 +730,26 @@ ssetup_ntlmssp_authenticate: } else ascii_ssetup_strings(&bcc_ptr, ses, nls_cp); } else if (type == NTLMv2) { - char *v2_sess_key = - kmalloc(sizeof(struct ntlmv2_resp), GFP_KERNEL); - - /* BB FIXME change all users of v2_sess_key to - struct ntlmv2_resp */ - - if (v2_sess_key == NULL) { - rc = -ENOMEM; - goto ssetup_exit; - } - pSMB->req_no_secext.Capabilities = cpu_to_le32(capabilities); /* LM2 password would be here if we supported it */ pSMB->req_no_secext.CaseInsensitivePasswordLength = 0; - /* cpu_to_le16(LM2_SESS_KEY_SIZE); */ - /* calculate session key */ - rc = setup_ntlmv2_rsp(ses, v2_sess_key, nls_cp); + /* calculate nlmv2 response and session key */ + rc = setup_ntlmv2_rsp(ses, nls_cp); if (rc) { cERROR(1, "Error %d during NTLMv2 authentication", rc); - kfree(v2_sess_key); goto ssetup_exit; } - memcpy(bcc_ptr, (char *)v2_sess_key, - sizeof(struct ntlmv2_resp)); - bcc_ptr += sizeof(struct ntlmv2_resp); - kfree(v2_sess_key); + memcpy(bcc_ptr, ses->auth_key.response + CIFS_SESS_KEY_SIZE, + ses->auth_key.len - CIFS_SESS_KEY_SIZE); + bcc_ptr += ses->auth_key.len - CIFS_SESS_KEY_SIZE; + /* set case sensitive password length after tilen may get * assigned, tilen is 0 otherwise. */ pSMB->req_no_secext.CaseSensitivePasswordLength = - cpu_to_le16(sizeof(struct ntlmv2_resp) + ses->tilen); - if (ses->tilen > 0) { - memcpy(bcc_ptr, ses->tiblob, ses->tilen); - bcc_ptr += ses->tilen; - /* we never did allocate ses->domainName to free */ - kfree(ses->tiblob); - ses->tiblob = NULL; - ses->tilen = 0; - } + cpu_to_le16(ses->auth_key.len - CIFS_SESS_KEY_SIZE); if (ses->capabilities & CAP_UNICODE) { if (iov[0].iov_len % 2) { @@ -768,6 +762,7 @@ ssetup_ntlmssp_authenticate: } else if (type == Kerberos) { #ifdef CONFIG_CIFS_UPCALL struct cifs_spnego_msg *msg; + spnego_key = cifs_get_spnego_key(ses); if (IS_ERR(spnego_key)) { rc = PTR_ERR(spnego_key); @@ -785,16 +780,17 @@ ssetup_ntlmssp_authenticate: rc = -EKEYREJECTED; goto ssetup_exit; } - /* bail out if key is too long */ - if (msg->sesskey_len > - sizeof(ses->auth_key.data.krb5)) { - cERROR(1, "Kerberos signing key too long (%u bytes)", - msg->sesskey_len); - rc = -EOVERFLOW; + + ses->auth_key.response = kmalloc(msg->sesskey_len, GFP_KERNEL); + if (!ses->auth_key.response) { + cERROR(1, "Kerberos can't allocate (%u bytes) memory", + msg->sesskey_len); + rc = -ENOMEM; goto ssetup_exit; } + memcpy(ses->auth_key.response, msg->data, msg->sesskey_len); ses->auth_key.len = msg->sesskey_len; - memcpy(ses->auth_key.data.krb5, msg->data, msg->sesskey_len); + pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; capabilities |= CAP_EXTENDED_SECURITY; pSMB->req.Capabilities = cpu_to_le32(capabilities); @@ -897,8 +893,6 @@ ssetup_ntlmssp_authenticate: CIFS_STD_OP /* not long */ | CIFS_LOG_ERROR); /* SMB request buf freed in SendReceive2 */ - cFYI(1, "ssetup rc from sendrecv2 is %d", rc); - pSMB = (SESSION_SETUP_ANDX *)iov[0].iov_base; smb_buf = (struct smb_hdr *)iov[0].iov_base; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index a66c91e..e0588cd 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -543,7 +543,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))) { rc = cifs_verify_signature(midQ->resp_buf, - &ses->server->session_key, + ses->server, midQ->sequence_number+1); if (rc) { cERROR(1, "Unexpected SMB signature"); @@ -731,7 +731,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))) { rc = cifs_verify_signature(out_buf, - &ses->server->session_key, + ses->server, midQ->sequence_number+1); if (rc) { cERROR(1, "Unexpected SMB signature"); @@ -981,7 +981,7 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon, (ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))) { rc = cifs_verify_signature(out_buf, - &ses->server->session_key, + ses->server, midQ->sequence_number+1); if (rc) { cERROR(1, "Unexpected SMB signature"); diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 7993b96..5ea57c8 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -306,16 +306,16 @@ static int coda_statfs(struct dentry *dentry, struct kstatfs *buf) /* init_coda: used by filesystems.c to register coda */ -static int coda_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *coda_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, coda_fill_super, mnt); + return mount_nodev(fs_type, flags, data, coda_fill_super); } struct file_system_type coda_fs_type = { .owner = THIS_MODULE, .name = "coda", - .get_sb = coda_get_sb, + .mount = coda_mount, .kill_sb = kill_anon_super, .fs_flags = FS_BINARY_MOUNTDATA, }; diff --git a/fs/compat.c b/fs/compat.c index 52cfeb6..c580c32 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -49,6 +49,7 @@ #include <linux/eventpoll.h> #include <linux/fs_struct.h> #include <linux/slab.h> +#include <linux/pagemap.h> #include <asm/uaccess.h> #include <asm/mmu_context.h> @@ -606,14 +607,14 @@ ssize_t compat_rw_copy_check_uvector(int type, /* * Single unix specification: * We should -EINVAL if an element length is not >= 0 and fitting an - * ssize_t. The total length is fitting an ssize_t + * ssize_t. * - * Be careful here because iov_len is a size_t not an ssize_t + * In Linux, the total length is limited to MAX_RW_COUNT, there is + * no overflow possibility. */ tot_len = 0; ret = -EINVAL; for (seg = 0; seg < nr_segs; seg++) { - compat_ssize_t tmp = tot_len; compat_uptr_t buf; compat_ssize_t len; @@ -624,13 +625,13 @@ ssize_t compat_rw_copy_check_uvector(int type, } if (len < 0) /* size_t not fitting in compat_ssize_t .. */ goto out; - tot_len += len; - if (tot_len < tmp) /* maths overflow on the compat_ssize_t */ - goto out; if (!access_ok(vrfy_dir(type), compat_ptr(buf), len)) { ret = -EFAULT; goto out; } + if (len > MAX_RW_COUNT - tot_len) + len = MAX_RW_COUNT - tot_len; + tot_len += len; iov->iov_base = compat_ptr(buf); iov->iov_len = (compat_size_t) len; uvector++; diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index 8c8d642..7d3607f 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -104,16 +104,16 @@ static int configfs_fill_super(struct super_block *sb, void *data, int silent) return 0; } -static int configfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *configfs_do_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_single(fs_type, flags, data, configfs_fill_super, mnt); + return mount_single(fs_type, flags, data, configfs_fill_super); } static struct file_system_type configfs_fs_type = { .owner = THIS_MODULE, .name = "configfs", - .get_sb = configfs_get_sb, + .mount = configfs_do_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 1e7a330..32fd5fe 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -533,17 +533,16 @@ static const struct super_operations cramfs_ops = { .statfs = cramfs_statfs, }; -static int cramfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *cramfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, cramfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, cramfs_fill_super); } static struct file_system_type cramfs_fs_type = { .owner = THIS_MODULE, .name = "cramfs", - .get_sb = cramfs_get_sb, + .mount = cramfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index a4ed838..37a8ca7 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -135,17 +135,17 @@ static int debug_fill_super(struct super_block *sb, void *data, int silent) return simple_fill_super(sb, DEBUGFS_MAGIC, debug_files); } -static int debug_get_sb(struct file_system_type *fs_type, +static struct dentry *debug_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_single(fs_type, flags, data, debug_fill_super, mnt); + return mount_single(fs_type, flags, data, debug_fill_super); } static struct file_system_type debug_fs_type = { .owner = THIS_MODULE, .name = "debugfs", - .get_sb = debug_get_sb, + .mount = debug_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 8b3ffd5..1bb547c 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -331,7 +331,7 @@ static int compare_init_pts_sb(struct super_block *s, void *p) } /* - * devpts_get_sb() + * devpts_mount() * * If the '-o newinstance' mount option was specified, mount a new * (private) instance of devpts. PTYs created in this instance are @@ -345,20 +345,20 @@ static int compare_init_pts_sb(struct super_block *s, void *p) * semantics in devpts while preserving backward compatibility of the * current 'single-namespace' semantics. i.e all mounts of devpts * without the 'newinstance' mount option should bind to the initial - * kernel mount, like get_sb_single(). + * kernel mount, like mount_single(). * * Mounts with 'newinstance' option create a new, private namespace. * * NOTE: * - * For single-mount semantics, devpts cannot use get_sb_single(), - * because get_sb_single()/sget() find and use the super-block from + * For single-mount semantics, devpts cannot use mount_single(), + * because mount_single()/sget() find and use the super-block from * the most recent mount of devpts. But that recent mount may be a - * 'newinstance' mount and get_sb_single() would pick the newinstance + * 'newinstance' mount and mount_single() would pick the newinstance * super-block instead of the initial super-block. */ -static int devpts_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *devpts_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { int error; struct pts_mount_opts opts; @@ -366,7 +366,7 @@ static int devpts_get_sb(struct file_system_type *fs_type, error = parse_mount_options(data, PARSE_MOUNT, &opts); if (error) - return error; + return ERR_PTR(error); if (opts.newinstance) s = sget(fs_type, NULL, set_anon_super, NULL); @@ -374,7 +374,7 @@ static int devpts_get_sb(struct file_system_type *fs_type, s = sget(fs_type, compare_init_pts_sb, set_anon_super, NULL); if (IS_ERR(s)) - return PTR_ERR(s); + return ERR_CAST(s); if (!s->s_root) { s->s_flags = flags; @@ -390,13 +390,11 @@ static int devpts_get_sb(struct file_system_type *fs_type, if (error) goto out_undo_sget; - simple_set_mnt(mnt, s); - - return 0; + return dget(s->s_root); out_undo_sget: deactivate_locked_super(s); - return error; + return ERR_PTR(error); } #else @@ -404,10 +402,10 @@ out_undo_sget: * This supports only the legacy single-instance semantics (no * multiple-instance semantics) */ -static int devpts_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *devpts_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { - return get_sb_single(fs_type, flags, data, devpts_fill_super, mnt); + return mount_single(fs_type, flags, data, devpts_fill_super); } #endif @@ -421,7 +419,7 @@ static void devpts_kill_sb(struct super_block *sb) static struct file_system_type devpts_fs_type = { .name = "devpts", - .get_sb = devpts_get_sb, + .mount = devpts_mount, .kill_sb = devpts_kill_sb, }; diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 40186b9..413a3c4 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -377,6 +377,7 @@ struct ecryptfs_mount_crypt_stat { #define ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES 0x00000010 #define ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK 0x00000020 #define ECRYPTFS_GLOBAL_ENCFN_USE_FEK 0x00000040 +#define ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY 0x00000080 u32 flags; struct list_head global_auth_tok_list; struct mutex global_auth_tok_list_mutex; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 3fbc942..9d1a22d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -32,6 +32,7 @@ #include <linux/crypto.h> #include <linux/fs_stack.h> #include <linux/slab.h> +#include <linux/xattr.h> #include <asm/unaligned.h> #include "ecryptfs_kernel.h" @@ -70,15 +71,19 @@ ecryptfs_create_underlying_file(struct inode *lower_dir_inode, struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); struct dentry *dentry_save; struct vfsmount *vfsmount_save; + unsigned int flags_save; int rc; dentry_save = nd->path.dentry; vfsmount_save = nd->path.mnt; + flags_save = nd->flags; nd->path.dentry = lower_dentry; nd->path.mnt = lower_mnt; + nd->flags &= ~LOOKUP_OPEN; rc = vfs_create(lower_dir_inode, lower_dentry, mode, nd); nd->path.dentry = dentry_save; nd->path.mnt = vfsmount_save; + nd->flags = flags_save; return rc; } @@ -1108,10 +1113,8 @@ ecryptfs_setxattr(struct dentry *dentry, const char *name, const void *value, rc = -EOPNOTSUPP; goto out; } - mutex_lock(&lower_dentry->d_inode->i_mutex); - rc = lower_dentry->d_inode->i_op->setxattr(lower_dentry, name, value, - size, flags); - mutex_unlock(&lower_dentry->d_inode->i_mutex); + + rc = vfs_setxattr(lower_dentry, name, value, size, flags); out: return rc; } diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 73811cf..b1f6858 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -446,6 +446,7 @@ out: */ static int ecryptfs_find_auth_tok_for_sig( + struct key **auth_tok_key, struct ecryptfs_auth_tok **auth_tok, struct ecryptfs_mount_crypt_stat *mount_crypt_stat, char *sig) @@ -453,12 +454,21 @@ ecryptfs_find_auth_tok_for_sig( struct ecryptfs_global_auth_tok *global_auth_tok; int rc = 0; + (*auth_tok_key) = NULL; (*auth_tok) = NULL; if (ecryptfs_find_global_auth_tok_for_sig(&global_auth_tok, mount_crypt_stat, sig)) { - struct key *auth_tok_key; - rc = ecryptfs_keyring_auth_tok_for_sig(&auth_tok_key, auth_tok, + /* if the flag ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY is set in the + * mount_crypt_stat structure, we prevent to use auth toks that + * are not inserted through the ecryptfs_add_global_auth_tok + * function. + */ + if (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY) + return -EINVAL; + + rc = ecryptfs_keyring_auth_tok_for_sig(auth_tok_key, auth_tok, sig); } else (*auth_tok) = global_auth_tok->global_auth_tok; @@ -509,6 +519,7 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, char *filename, size_t filename_size) { struct ecryptfs_write_tag_70_packet_silly_stack *s; + struct key *auth_tok_key = NULL; int rc = 0; s = kmalloc(sizeof(*s), GFP_KERNEL); @@ -606,6 +617,7 @@ ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, } dest[s->i++] = s->cipher_code; rc = ecryptfs_find_auth_tok_for_sig( + &auth_tok_key, &s->auth_tok, mount_crypt_stat, mount_crypt_stat->global_default_fnek_sig); if (rc) { @@ -753,6 +765,8 @@ out_free_unlock: out_unlock: mutex_unlock(s->tfm_mutex); out: + if (auth_tok_key) + key_put(auth_tok_key); kfree(s); return rc; } @@ -798,6 +812,7 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, char *data, size_t max_packet_size) { struct ecryptfs_parse_tag_70_packet_silly_stack *s; + struct key *auth_tok_key = NULL; int rc = 0; (*packet_size) = 0; @@ -910,7 +925,8 @@ ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, * >= ECRYPTFS_MAX_IV_BYTES. */ memset(s->iv, 0, ECRYPTFS_MAX_IV_BYTES); s->desc.info = s->iv; - rc = ecryptfs_find_auth_tok_for_sig(&s->auth_tok, mount_crypt_stat, + rc = ecryptfs_find_auth_tok_for_sig(&auth_tok_key, + &s->auth_tok, mount_crypt_stat, s->fnek_sig_hex); if (rc) { printk(KERN_ERR "%s: Error attempting to find auth tok for " @@ -986,6 +1002,8 @@ out: (*filename_size) = 0; (*filename) = NULL; } + if (auth_tok_key) + key_put(auth_tok_key); kfree(s); return rc; } @@ -1557,14 +1575,19 @@ int ecryptfs_keyring_auth_tok_for_sig(struct key **auth_tok_key, ECRYPTFS_VERSION_MAJOR, ECRYPTFS_VERSION_MINOR); rc = -EINVAL; - goto out; + goto out_release_key; } if ((*auth_tok)->token_type != ECRYPTFS_PASSWORD && (*auth_tok)->token_type != ECRYPTFS_PRIVATE_KEY) { printk(KERN_ERR "Invalid auth_tok structure " "returned from key query\n"); rc = -EINVAL; - goto out; + goto out_release_key; + } +out_release_key: + if (rc) { + key_put(*auth_tok_key); + (*auth_tok_key) = NULL; } out: return rc; @@ -1688,6 +1711,7 @@ int ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, struct ecryptfs_auth_tok_list_item *auth_tok_list_item; size_t tag_11_contents_size; size_t tag_11_packet_size; + struct key *auth_tok_key = NULL; int rc = 0; INIT_LIST_HEAD(&auth_tok_list); @@ -1784,6 +1808,10 @@ int ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, * just one will be sufficient to decrypt to get the FEK. */ find_next_matching_auth_tok: found_auth_tok = 0; + if (auth_tok_key) { + key_put(auth_tok_key); + auth_tok_key = NULL; + } list_for_each_entry(auth_tok_list_item, &auth_tok_list, list) { candidate_auth_tok = &auth_tok_list_item->auth_tok; if (unlikely(ecryptfs_verbosity > 0)) { @@ -1800,10 +1828,11 @@ find_next_matching_auth_tok: rc = -EINVAL; goto out_wipe_list; } - ecryptfs_find_auth_tok_for_sig(&matching_auth_tok, + rc = ecryptfs_find_auth_tok_for_sig(&auth_tok_key, + &matching_auth_tok, crypt_stat->mount_crypt_stat, candidate_auth_tok_sig); - if (matching_auth_tok) { + if (!rc) { found_auth_tok = 1; goto found_matching_auth_tok; } @@ -1866,6 +1895,8 @@ found_matching_auth_tok: out_wipe_list: wipe_auth_tok_list(&auth_tok_list); out: + if (auth_tok_key) + key_put(auth_tok_key); return rc; } diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index cbd4e18..a9dbd62 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -208,7 +208,8 @@ enum { ecryptfs_opt_sig, ecryptfs_opt_ecryptfs_sig, ecryptfs_opt_passthrough, ecryptfs_opt_xattr_metadata, ecryptfs_opt_encrypted_view, ecryptfs_opt_fnek_sig, ecryptfs_opt_fn_cipher, ecryptfs_opt_fn_cipher_key_bytes, - ecryptfs_opt_unlink_sigs, ecryptfs_opt_err }; + ecryptfs_opt_unlink_sigs, ecryptfs_opt_mount_auth_tok_only, + ecryptfs_opt_err }; static const match_table_t tokens = { {ecryptfs_opt_sig, "sig=%s"}, @@ -223,6 +224,7 @@ static const match_table_t tokens = { {ecryptfs_opt_fn_cipher, "ecryptfs_fn_cipher=%s"}, {ecryptfs_opt_fn_cipher_key_bytes, "ecryptfs_fn_key_bytes=%u"}, {ecryptfs_opt_unlink_sigs, "ecryptfs_unlink_sigs"}, + {ecryptfs_opt_mount_auth_tok_only, "ecryptfs_mount_auth_tok_only"}, {ecryptfs_opt_err, NULL} }; @@ -406,6 +408,10 @@ static int ecryptfs_parse_options(struct ecryptfs_sb_info *sbi, char *options) case ecryptfs_opt_unlink_sigs: mount_crypt_stat->flags |= ECRYPTFS_UNLINK_SIGS; break; + case ecryptfs_opt_mount_auth_tok_only: + mount_crypt_stat->flags |= + ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY; + break; case ecryptfs_opt_err: default: printk(KERN_WARNING @@ -540,9 +546,8 @@ out: * ecryptfs_interpose to perform most of the linking * ecryptfs_interpose(): links the lower filesystem into ecryptfs (inode.c) */ -static int ecryptfs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *raw_data, - struct vfsmount *mnt) +static struct dentry *ecryptfs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data) { struct super_block *s; struct ecryptfs_sb_info *sbi; @@ -607,8 +612,7 @@ static int ecryptfs_get_sb(struct file_system_type *fs_type, int flags, err = "Reading sb failed"; goto out; } - simple_set_mnt(mnt, s); - return 0; + return dget(s->s_root); out: if (sbi) { @@ -616,7 +620,7 @@ out: kmem_cache_free(ecryptfs_sb_info_cache, sbi); } printk(KERN_ERR "%s; rc = [%d]\n", err, rc); - return rc; + return ERR_PTR(rc); } /** @@ -639,7 +643,7 @@ static void ecryptfs_kill_block_super(struct super_block *sb) static struct file_system_type ecryptfs_fs_type = { .owner = THIS_MODULE, .name = "ecryptfs", - .get_sb = ecryptfs_get_sb, + .mount = ecryptfs_mount, .kill_sb = ecryptfs_kill_block_super, .fs_flags = 0 }; diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index f7fc286..2537323 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -180,6 +180,8 @@ static int ecryptfs_show_options(struct seq_file *m, struct vfsmount *mnt) seq_printf(m, ",ecryptfs_encrypted_view"); if (mount_crypt_stat->flags & ECRYPTFS_UNLINK_SIGS) seq_printf(m, ",ecryptfs_unlink_sigs"); + if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_MOUNT_AUTH_TOK_ONLY) + seq_printf(m, ",ecryptfs_mount_auth_tok_only"); return 0; } diff --git a/fs/efs/super.c b/fs/efs/super.c index f049428..5073a07 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -20,16 +20,16 @@ static int efs_statfs(struct dentry *dentry, struct kstatfs *buf); static int efs_fill_super(struct super_block *s, void *d, int silent); -static int efs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *efs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, efs_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, efs_fill_super); } static struct file_system_type efs_fs_type = { .owner = THIS_MODULE, .name = "efs", - .get_sb = efs_get_sb, + .mount = efs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 047e92f..79c3ae6 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -659,19 +659,19 @@ free_bdi: /* * Set up the superblock (calls exofs_fill_super eventually) */ -static int exofs_get_sb(struct file_system_type *type, +static struct dentry *exofs_mount(struct file_system_type *type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { struct exofs_mountopt opts; int ret; ret = parse_options(data, &opts); if (ret) - return ret; + return ERR_PTR(ret); opts.dev_name = dev_name; - return get_sb_nodev(type, flags, &opts, exofs_fill_super, mnt); + return mount_nodev(type, flags, &opts, exofs_fill_super); } /* @@ -809,7 +809,7 @@ static const struct export_operations exofs_export_ops = { static struct file_system_type exofs_type = { .owner = THIS_MODULE, .name = "exofs", - .get_sb = exofs_get_sb, + .mount = exofs_mount, .kill_sb = generic_shutdown_super, }; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 09013206..d89e0b6 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1356,10 +1356,10 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) return 0; } -static int ext2_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ext2_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ext2_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, ext2_fill_super); } #ifdef CONFIG_QUOTA @@ -1473,7 +1473,7 @@ out: static struct file_system_type ext2_fs_type = { .owner = THIS_MODULE, .name = "ext2", - .get_sb = ext2_get_sb, + .mount = ext2_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ext3/super.c b/fs/ext3/super.c index db87413..2fedaf8 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -3020,16 +3020,16 @@ out: #endif -static int ext3_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ext3_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ext3_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, ext3_fill_super); } static struct file_system_type ext3_fs_type = { .owner = THIS_MODULE, .name = "ext3", - .get_sb = ext3_get_sb, + .mount = ext3_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 0348ce0..40131b7 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -73,8 +73,8 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf); static int ext4_unfreeze(struct super_block *sb); static void ext4_write_super(struct super_block *sb); static int ext4_freeze(struct super_block *sb); -static int ext4_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt); +static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data); static void ext4_destroy_lazyinit_thread(void); static void ext4_unregister_li_request(struct super_block *sb); @@ -82,7 +82,7 @@ static void ext4_unregister_li_request(struct super_block *sb); static struct file_system_type ext3_fs_type = { .owner = THIS_MODULE, .name = "ext3", - .get_sb = ext4_get_sb, + .mount = ext4_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -4667,17 +4667,17 @@ out: #endif -static int ext4_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ext4_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super,mnt); + return mount_bdev(fs_type, flags, dev_name, data, ext4_fill_super); } #if !defined(CONFIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23) static struct file_system_type ext2_fs_type = { .owner = THIS_MODULE, .name = "ext2", - .get_sb = ext4_get_sb, + .mount = ext4_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -4722,7 +4722,7 @@ static inline void unregister_as_ext3(void) { } static struct file_system_type ext4_fs_type = { .owner = THIS_MODULE, .name = "ext4", - .get_sb = ext4_get_sb, + .mount = ext4_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index bbca5c1..3345aab 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -675,18 +675,17 @@ static int msdos_fill_super(struct super_block *sb, void *data, int silent) return 0; } -static int msdos_get_sb(struct file_system_type *fs_type, +static struct dentry *msdos_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, msdos_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, msdos_fill_super); } static struct file_system_type msdos_fs_type = { .owner = THIS_MODULE, .name = "msdos", - .get_sb = msdos_get_sb, + .mount = msdos_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 6f0f6c9..b936703 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -1071,18 +1071,17 @@ static int vfat_fill_super(struct super_block *sb, void *data, int silent) return 0; } -static int vfat_get_sb(struct file_system_type *fs_type, +static struct dentry *vfat_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, vfat_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, vfat_fill_super); } static struct file_system_type vfat_fs_type = { .owner = THIS_MODULE, .name = "vfat", - .get_sb = vfat_get_sb, + .mount = vfat_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c index 71b0148..9d1c995 100644 --- a/fs/freevxfs/vxfs_super.c +++ b/fs/freevxfs/vxfs_super.c @@ -246,17 +246,16 @@ out: /* * The usual module blurb. */ -static int vxfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *vxfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, vxfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, vxfs_fill_super); } static struct file_system_type vxfs_fs_type = { .owner = THIS_MODULE, .name = "vxfs", - .get_sb = vxfs_get_sb, + .mount = vxfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index aed881a..3d06ccc 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -707,6 +707,17 @@ get_next_work_item(struct backing_dev_info *bdi) return work; } +/* + * Add in the number of potentially dirty inodes, because each inode + * write can dirty pagecache in the underlying blockdev. + */ +static unsigned long get_nr_dirty_pages(void) +{ + return global_page_state(NR_FILE_DIRTY) + + global_page_state(NR_UNSTABLE_NFS) + + get_nr_dirty_inodes(); +} + static long wb_check_old_data_flush(struct bdi_writeback *wb) { unsigned long expired; @@ -724,13 +735,7 @@ static long wb_check_old_data_flush(struct bdi_writeback *wb) return 0; wb->last_old_flush = jiffies; - /* - * Add in the number of potentially dirty inodes, because each inode - * write can dirty pagecache in the underlying blockdev. - */ - nr_pages = global_page_state(NR_FILE_DIRTY) + - global_page_state(NR_UNSTABLE_NFS) + - get_nr_dirty_inodes(); + nr_pages = get_nr_dirty_pages(); if (nr_pages) { struct wb_writeback_work work = { @@ -1076,32 +1081,42 @@ static void wait_sb_inodes(struct super_block *sb) } /** - * writeback_inodes_sb - writeback dirty inodes from given super_block + * writeback_inodes_sb_nr - writeback dirty inodes from given super_block * @sb: the superblock + * @nr: the number of pages to write * * Start writeback on some inodes on this super_block. No guarantees are made * on how many (if any) will be written, and this function does not wait - * for IO completion of submitted IO. The number of pages submitted is - * returned. + * for IO completion of submitted IO. */ -void writeback_inodes_sb(struct super_block *sb) +void writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr) { - unsigned long nr_dirty = global_page_state(NR_FILE_DIRTY); - unsigned long nr_unstable = global_page_state(NR_UNSTABLE_NFS); DECLARE_COMPLETION_ONSTACK(done); struct wb_writeback_work work = { .sb = sb, .sync_mode = WB_SYNC_NONE, .done = &done, + .nr_pages = nr, }; WARN_ON(!rwsem_is_locked(&sb->s_umount)); - - work.nr_pages = nr_dirty + nr_unstable + get_nr_dirty_inodes(); - bdi_queue_work(sb->s_bdi, &work); wait_for_completion(&done); } +EXPORT_SYMBOL(writeback_inodes_sb_nr); + +/** + * writeback_inodes_sb - writeback dirty inodes from given super_block + * @sb: the superblock + * + * Start writeback on some inodes on this super_block. No guarantees are made + * on how many (if any) will be written, and this function does not wait + * for IO completion of submitted IO. + */ +void writeback_inodes_sb(struct super_block *sb) +{ + return writeback_inodes_sb_nr(sb, get_nr_dirty_pages()); +} EXPORT_SYMBOL(writeback_inodes_sb); /** @@ -1124,6 +1139,27 @@ int writeback_inodes_sb_if_idle(struct super_block *sb) EXPORT_SYMBOL(writeback_inodes_sb_if_idle); /** + * writeback_inodes_sb_if_idle - start writeback if none underway + * @sb: the superblock + * @nr: the number of pages to write + * + * Invoke writeback_inodes_sb if no writeback is currently underway. + * Returns 1 if writeback was started, 0 if not. + */ +int writeback_inodes_sb_nr_if_idle(struct super_block *sb, + unsigned long nr) +{ + if (!writeback_in_progress(sb->s_bdi)) { + down_read(&sb->s_umount); + writeback_inodes_sb_nr(sb, nr); + up_read(&sb->s_umount); + return 1; + } else + return 0; +} +EXPORT_SYMBOL(writeback_inodes_sb_nr_if_idle); + +/** * sync_inodes_sb - sync sb inode pages * @sb: the superblock * diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 4eba076..85542a7 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -322,12 +322,10 @@ static int fuse_ctl_fill_super(struct super_block *sb, void *data, int silent) return 0; } -static int fuse_ctl_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *raw_data, - struct vfsmount *mnt) +static struct dentry *fuse_ctl_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data) { - return get_sb_single(fs_type, flags, raw_data, - fuse_ctl_fill_super, mnt); + return mount_single(fs_type, flags, raw_data, fuse_ctl_fill_super); } static void fuse_ctl_kill_sb(struct super_block *sb) @@ -346,7 +344,7 @@ static void fuse_ctl_kill_sb(struct super_block *sb) static struct file_system_type fuse_ctl_fs_type = { .owner = THIS_MODULE, .name = "fusectl", - .get_sb = fuse_ctl_get_sb, + .mount = fuse_ctl_mount, .kill_sb = fuse_ctl_kill_sb, }; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index da9e6e1..cfce3ad 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -1041,11 +1041,11 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) return err; } -static int fuse_get_sb(struct file_system_type *fs_type, +static struct dentry *fuse_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *raw_data, struct vfsmount *mnt) + void *raw_data) { - return get_sb_nodev(fs_type, flags, raw_data, fuse_fill_super, mnt); + return mount_nodev(fs_type, flags, raw_data, fuse_fill_super); } static void fuse_kill_sb_anon(struct super_block *sb) @@ -1065,17 +1065,16 @@ static struct file_system_type fuse_fs_type = { .owner = THIS_MODULE, .name = "fuse", .fs_flags = FS_HAS_SUBTYPE, - .get_sb = fuse_get_sb, + .mount = fuse_mount, .kill_sb = fuse_kill_sb_anon, }; #ifdef CONFIG_BLOCK -static int fuse_get_sb_blk(struct file_system_type *fs_type, +static struct dentry *fuse_mount_blk(struct file_system_type *fs_type, int flags, const char *dev_name, - void *raw_data, struct vfsmount *mnt) + void *raw_data) { - return get_sb_bdev(fs_type, flags, dev_name, raw_data, fuse_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, raw_data, fuse_fill_super); } static void fuse_kill_sb_blk(struct super_block *sb) @@ -1094,7 +1093,7 @@ static void fuse_kill_sb_blk(struct super_block *sb) static struct file_system_type fuseblk_fs_type = { .owner = THIS_MODULE, .name = "fuseblk", - .get_sb = fuse_get_sb_blk, + .mount = fuse_mount_blk, .kill_sb = fuse_kill_sb_blk, .fs_flags = FS_REQUIRES_DEV | FS_HAS_SUBTYPE, }; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index cade1ac..3eb1393 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1250,12 +1250,11 @@ static int test_gfs2_super(struct super_block *s, void *ptr) } /** - * gfs2_get_sb - Get the GFS2 superblock + * gfs2_mount - Get the GFS2 superblock * @fs_type: The GFS2 filesystem type * @flags: Mount flags * @dev_name: The name of the device * @data: The mount arguments - * @mnt: The vfsmnt for this mount * * Q. Why not use get_sb_bdev() ? * A. We need to select one of two root directories to mount, independent @@ -1264,8 +1263,8 @@ static int test_gfs2_super(struct super_block *s, void *ptr) * Returns: 0 or -ve on error */ -static int gfs2_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *gfs2_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { struct block_device *bdev; struct super_block *s; @@ -1279,7 +1278,7 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags, bdev = open_bdev_exclusive(dev_name, mode, fs_type); if (IS_ERR(bdev)) - return PTR_ERR(bdev); + return ERR_CAST(bdev); /* * once the super is inserted into the list by sget, s_umount @@ -1298,6 +1297,9 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags, if (IS_ERR(s)) goto error_bdev; + if (s->s_root) + close_bdev_exclusive(bdev, mode); + memset(&args, 0, sizeof(args)); args.ar_quota = GFS2_QUOTA_DEFAULT; args.ar_data = GFS2_DATA_DEFAULT; @@ -1309,17 +1311,13 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags, error = gfs2_mount_args(&args, data); if (error) { printk(KERN_WARNING "GFS2: can't parse mount arguments\n"); - if (s->s_root) - goto error_super; - deactivate_locked_super(s); - return error; + goto error_super; } if (s->s_root) { error = -EBUSY; if ((flags ^ s->s_flags) & MS_RDONLY) goto error_super; - close_bdev_exclusive(bdev, mode); } else { char b[BDEVNAME_SIZE]; @@ -1328,27 +1326,24 @@ static int gfs2_get_sb(struct file_system_type *fs_type, int flags, strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); sb_set_blocksize(s, block_size(bdev)); error = fill_super(s, &args, flags & MS_SILENT ? 1 : 0); - if (error) { - deactivate_locked_super(s); - return error; - } + if (error) + goto error_super; s->s_flags |= MS_ACTIVE; bdev->bd_super = s; } sdp = s->s_fs_info; - mnt->mnt_sb = s; if (args.ar_meta) - mnt->mnt_root = dget(sdp->sd_master_dir); + return dget(sdp->sd_master_dir); else - mnt->mnt_root = dget(sdp->sd_root_dir); - return 0; + return dget(sdp->sd_root_dir); error_super: deactivate_locked_super(s); + return ERR_PTR(error); error_bdev: close_bdev_exclusive(bdev, mode); - return error; + return ERR_PTR(error); } static int set_meta_super(struct super_block *s, void *ptr) @@ -1356,8 +1351,8 @@ static int set_meta_super(struct super_block *s, void *ptr) return -EINVAL; } -static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *gfs2_mount_meta(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { struct super_block *s; struct gfs2_sbd *sdp; @@ -1368,23 +1363,21 @@ static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, if (error) { printk(KERN_WARNING "GFS2: path_lookup on %s returned error %d\n", dev_name, error); - return error; + return ERR_PTR(error); } s = sget(&gfs2_fs_type, test_gfs2_super, set_meta_super, path.dentry->d_inode->i_sb->s_bdev); path_put(&path); if (IS_ERR(s)) { printk(KERN_WARNING "GFS2: gfs2 mount does not exist\n"); - return PTR_ERR(s); + return ERR_CAST(s); } if ((flags ^ s->s_flags) & MS_RDONLY) { deactivate_locked_super(s); - return -EBUSY; + return ERR_PTR(-EBUSY); } sdp = s->s_fs_info; - mnt->mnt_sb = s; - mnt->mnt_root = dget(sdp->sd_master_dir); - return 0; + return dget(sdp->sd_master_dir); } static void gfs2_kill_sb(struct super_block *sb) @@ -1410,7 +1403,7 @@ static void gfs2_kill_sb(struct super_block *sb) struct file_system_type gfs2_fs_type = { .name = "gfs2", .fs_flags = FS_REQUIRES_DEV, - .get_sb = gfs2_get_sb, + .mount = gfs2_mount, .kill_sb = gfs2_kill_sb, .owner = THIS_MODULE, }; @@ -1418,7 +1411,7 @@ struct file_system_type gfs2_fs_type = { struct file_system_type gfs2meta_fs_type = { .name = "gfs2meta", .fs_flags = FS_REQUIRES_DEV, - .get_sb = gfs2_get_sb_meta, + .mount = gfs2_mount_meta, .owner = THIS_MODULE, }; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 6ee1586..4824c27 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -441,17 +441,16 @@ bail: return res; } -static int hfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *hfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, hfs_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, hfs_fill_super); } static struct file_system_type hfs_fs_type = { .owner = THIS_MODULE, .name = "hfs", - .get_sb = hfs_get_sb, + .mount = hfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 9a88d75..52cc746 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -495,18 +495,16 @@ static void hfsplus_destroy_inode(struct inode *inode) #define HFSPLUS_INODE_SIZE sizeof(struct hfsplus_inode_info) -static int hfsplus_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *hfsplus_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, hfsplus_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, hfsplus_fill_super); } static struct file_system_type hfsplus_fs_type = { .owner = THIS_MODULE, .name = "hfsplus", - .get_sb = hfsplus_get_sb, + .mount = hfsplus_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index cd7c939..2c0f148 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -962,11 +962,11 @@ out: return err; } -static int hostfs_read_sb(struct file_system_type *type, +static struct dentry *hostfs_read_sb(struct file_system_type *type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_nodev(type, flags, data, hostfs_fill_sb_common, mnt); + return mount_nodev(type, flags, data, hostfs_fill_sb_common); } static void hostfs_kill_sb(struct super_block *s) @@ -978,7 +978,7 @@ static void hostfs_kill_sb(struct super_block *s) static struct file_system_type hostfs_type = { .owner = THIS_MODULE, .name = "hostfs", - .get_sb = hostfs_read_sb, + .mount = hostfs_read_sb, .kill_sb = hostfs_kill_sb, .fs_flags = 0, }; diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 18e1d45..6c5f015 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -686,17 +686,16 @@ bail0: return -EINVAL; } -static int hpfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *hpfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, hpfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, hpfs_fill_super); } static struct file_system_type hpfs_fs_type = { .owner = THIS_MODULE, .name = "hpfs", - .get_sb = hpfs_get_sb, + .mount = hpfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 4e2a45e..f702b5f 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -748,17 +748,17 @@ static int hppfs_fill_super(struct super_block *sb, void *d, int silent) return(err); } -static int hppfs_read_super(struct file_system_type *type, +static struct dentry *hppfs_read_super(struct file_system_type *type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_nodev(type, flags, data, hppfs_fill_super, mnt); + return mount_nodev(type, flags, data, hppfs_fill_super); } static struct file_system_type hppfs_type = { .owner = THIS_MODULE, .name = "hppfs", - .get_sb = hppfs_read_super, + .mount = hppfs_read_super, .kill_sb = kill_anon_super, .fs_flags = 0, }; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index b14be3f..d6cfac1 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -896,15 +896,15 @@ void hugetlb_put_quota(struct address_space *mapping, long delta) } } -static int hugetlbfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *hugetlbfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, hugetlbfs_fill_super, mnt); + return mount_nodev(fs_type, flags, data, hugetlbfs_fill_super); } static struct file_system_type hugetlbfs_fs_type = { .name = "hugetlbfs", - .get_sb = hugetlbfs_get_sb, + .mount = hugetlbfs_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/internal.h b/fs/internal.h index ebad3b9..e43b9a4 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -106,5 +106,5 @@ extern void release_open_intent(struct nameidata *); * inode.c */ extern int get_nr_dirty_inodes(void); -extern int evict_inodes(struct super_block *); +extern void evict_inodes(struct super_block *); extern int invalidate_inodes(struct super_block *); diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 79cf7f61..bfdeb82 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -1507,17 +1507,16 @@ struct inode *isofs_iget(struct super_block *sb, return inode; } -static int isofs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *isofs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, isofs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, isofs_fill_super); } static struct file_system_type iso9660_fs_type = { .owner = THIS_MODULE, .name = "iso9660", - .get_sb = isofs_get_sb, + .mount = isofs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/jffs2/build.c b/fs/jffs2/build.c index a906f53..85c6be2 100644 --- a/fs/jffs2/build.c +++ b/fs/jffs2/build.c @@ -23,7 +23,7 @@ static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *, static inline struct jffs2_inode_cache * first_inode_chain(int *i, struct jffs2_sb_info *c) { - for (; *i < INOCACHE_HASHSIZE; (*i)++) { + for (; *i < c->inocache_hashsize; (*i)++) { if (c->inocache_list[*i]) return c->inocache_list[*i]; } diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c index 617a1e5..de42470 100644 --- a/fs/jffs2/compr.c +++ b/fs/jffs2/compr.c @@ -103,7 +103,7 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, spin_unlock(&jffs2_compressor_list_lock); *datalen = orig_slen; *cdatalen = orig_dlen; - compr_ret = this->compress(data_in, output_buf, datalen, cdatalen, NULL); + compr_ret = this->compress(data_in, output_buf, datalen, cdatalen); spin_lock(&jffs2_compressor_list_lock); this->usecount--; if (!compr_ret) { @@ -152,7 +152,7 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, spin_unlock(&jffs2_compressor_list_lock); *datalen = orig_slen; *cdatalen = orig_dlen; - compr_ret = this->compress(data_in, this->compr_buf, datalen, cdatalen, NULL); + compr_ret = this->compress(data_in, this->compr_buf, datalen, cdatalen); spin_lock(&jffs2_compressor_list_lock); this->usecount--; if (!compr_ret) { @@ -220,7 +220,7 @@ int jffs2_decompress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, if (comprtype == this->compr) { this->usecount++; spin_unlock(&jffs2_compressor_list_lock); - ret = this->decompress(cdata_in, data_out, cdatalen, datalen, NULL); + ret = this->decompress(cdata_in, data_out, cdatalen, datalen); spin_lock(&jffs2_compressor_list_lock); if (ret) { printk(KERN_WARNING "Decompressor \"%s\" returned %d\n", this->name, ret); diff --git a/fs/jffs2/compr.h b/fs/jffs2/compr.h index e471a91..13bb759 100644 --- a/fs/jffs2/compr.h +++ b/fs/jffs2/compr.h @@ -49,9 +49,9 @@ struct jffs2_compressor { char *name; char compr; /* JFFS2_COMPR_XXX */ int (*compress)(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *srclen, uint32_t *destlen, void *model); + uint32_t *srclen, uint32_t *destlen); int (*decompress)(unsigned char *cdata_in, unsigned char *data_out, - uint32_t cdatalen, uint32_t datalen, void *model); + uint32_t cdatalen, uint32_t datalen); int usecount; int disabled; /* if set the compressor won't compress */ unsigned char *compr_buf; /* used by size compr. mode */ diff --git a/fs/jffs2/compr_lzo.c b/fs/jffs2/compr_lzo.c index ed25ae7c..af186ee 100644 --- a/fs/jffs2/compr_lzo.c +++ b/fs/jffs2/compr_lzo.c @@ -42,7 +42,7 @@ static int __init alloc_workspace(void) } static int jffs2_lzo_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, void *model) + uint32_t *sourcelen, uint32_t *dstlen) { size_t compress_size; int ret; @@ -67,7 +67,7 @@ static int jffs2_lzo_compress(unsigned char *data_in, unsigned char *cpage_out, } static int jffs2_lzo_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t srclen, uint32_t destlen, void *model) + uint32_t srclen, uint32_t destlen) { size_t dl = destlen; int ret; diff --git a/fs/jffs2/compr_rtime.c b/fs/jffs2/compr_rtime.c index 9696ad9..16a5047 100644 --- a/fs/jffs2/compr_rtime.c +++ b/fs/jffs2/compr_rtime.c @@ -31,8 +31,7 @@ /* _compress returns the compressed size, -1 if bigger */ static int jffs2_rtime_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, - void *model) + uint32_t *sourcelen, uint32_t *dstlen) { short positions[256]; int outpos = 0; @@ -73,8 +72,7 @@ static int jffs2_rtime_compress(unsigned char *data_in, static int jffs2_rtime_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t srclen, uint32_t destlen, - void *model) + uint32_t srclen, uint32_t destlen) { short positions[256]; int outpos = 0; diff --git a/fs/jffs2/compr_rubin.c b/fs/jffs2/compr_rubin.c index a12b4f7..9e7cec8 100644 --- a/fs/jffs2/compr_rubin.c +++ b/fs/jffs2/compr_rubin.c @@ -298,7 +298,7 @@ static int rubin_do_compress(int bit_divider, int *bits, unsigned char *data_in, #if 0 /* _compress returns the compressed size, -1 if bigger */ int jffs2_rubinmips_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, void *model) + uint32_t *sourcelen, uint32_t *dstlen) { return rubin_do_compress(BIT_DIVIDER_MIPS, bits_mips, data_in, cpage_out, sourcelen, dstlen); @@ -306,8 +306,7 @@ int jffs2_rubinmips_compress(unsigned char *data_in, unsigned char *cpage_out, #endif static int jffs2_dynrubin_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, - void *model) + uint32_t *sourcelen, uint32_t *dstlen) { int bits[8]; unsigned char histo[256]; @@ -387,8 +386,7 @@ static void rubin_do_decompress(int bit_divider, int *bits, static int jffs2_rubinmips_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t sourcelen, uint32_t dstlen, - void *model) + uint32_t sourcelen, uint32_t dstlen) { rubin_do_decompress(BIT_DIVIDER_MIPS, bits_mips, data_in, cpage_out, sourcelen, dstlen); @@ -397,8 +395,7 @@ static int jffs2_rubinmips_decompress(unsigned char *data_in, static int jffs2_dynrubin_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t sourcelen, uint32_t dstlen, - void *model) + uint32_t sourcelen, uint32_t dstlen) { int bits[8]; int c; diff --git a/fs/jffs2/compr_zlib.c b/fs/jffs2/compr_zlib.c index 97fc45d..fd05a0b 100644 --- a/fs/jffs2/compr_zlib.c +++ b/fs/jffs2/compr_zlib.c @@ -68,8 +68,7 @@ static void free_workspaces(void) static int jffs2_zlib_compress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t *sourcelen, uint32_t *dstlen, - void *model) + uint32_t *sourcelen, uint32_t *dstlen) { int ret; @@ -136,8 +135,7 @@ static int jffs2_zlib_compress(unsigned char *data_in, static int jffs2_zlib_decompress(unsigned char *data_in, unsigned char *cpage_out, - uint32_t srclen, uint32_t destlen, - void *model) + uint32_t srclen, uint32_t destlen) { int ret; int wbits = MAX_WBITS; diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 79121aa..9297865 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -367,7 +367,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char } /* We use f->target field to store the target path. */ - f->target = kmalloc(targetlen + 1, GFP_KERNEL); + f->target = kmemdup(target, targetlen + 1, GFP_KERNEL); if (!f->target) { printk(KERN_WARNING "Can't allocate %d bytes of memory\n", targetlen + 1); mutex_unlock(&f->sem); @@ -376,7 +376,6 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char goto fail; } - memcpy(f->target, target, targetlen + 1); D1(printk(KERN_DEBUG "jffs2_symlink: symlink's target '%s' cached\n", (char *)f->target)); /* No data here. Only a metadata node, which will be diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index abac961..e513f19 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -151,7 +151,7 @@ int jffs2_erase_pending_blocks(struct jffs2_sb_info *c, int count) } /* Be nice */ - yield(); + cond_resched(); mutex_lock(&c->erase_free_sem); spin_lock(&c->erase_completion_lock); } diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index d9beb06..e896e67 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -474,6 +474,25 @@ struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_i return inode; } +static int calculate_inocache_hashsize(uint32_t flash_size) +{ + /* + * Pick a inocache hash size based on the size of the medium. + * Count how many megabytes we're dealing with, apply a hashsize twice + * that size, but rounding down to the usual big powers of 2. And keep + * to sensible bounds. + */ + + int size_mb = flash_size / 1024 / 1024; + int hashsize = (size_mb * 2) & ~0x3f; + + if (hashsize < INOCACHE_HASHSIZE_MIN) + return INOCACHE_HASHSIZE_MIN; + if (hashsize > INOCACHE_HASHSIZE_MAX) + return INOCACHE_HASHSIZE_MAX; + + return hashsize; +} int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) { @@ -520,7 +539,8 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) if (ret) return ret; - c->inocache_list = kcalloc(INOCACHE_HASHSIZE, sizeof(struct jffs2_inode_cache *), GFP_KERNEL); + c->inocache_hashsize = calculate_inocache_hashsize(c->flash_size); + c->inocache_list = kcalloc(c->inocache_hashsize, sizeof(struct jffs2_inode_cache *), GFP_KERNEL); if (!c->inocache_list) { ret = -ENOMEM; goto out_wbuf; diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c index 846a794..31dce61133 100644 --- a/fs/jffs2/gc.c +++ b/fs/jffs2/gc.c @@ -219,13 +219,14 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) if (!list_empty(&c->erase_complete_list) || !list_empty(&c->erase_pending_list)) { spin_unlock(&c->erase_completion_lock); + mutex_unlock(&c->alloc_sem); D1(printk(KERN_DEBUG "jffs2_garbage_collect_pass() erasing pending blocks\n")); - if (jffs2_erase_pending_blocks(c, 1)) { - mutex_unlock(&c->alloc_sem); + if (jffs2_erase_pending_blocks(c, 1)) return 0; - } + D1(printk(KERN_DEBUG "No progress from erasing blocks; doing GC anyway\n")); spin_lock(&c->erase_completion_lock); + mutex_lock(&c->alloc_sem); } /* First, work out which block we're garbage-collecting */ diff --git a/fs/jffs2/jffs2_fs_sb.h b/fs/jffs2/jffs2_fs_sb.h index 6784bc8..f864005 100644 --- a/fs/jffs2/jffs2_fs_sb.h +++ b/fs/jffs2/jffs2_fs_sb.h @@ -100,6 +100,7 @@ struct jffs2_sb_info { wait_queue_head_t erase_wait; /* For waiting for erases to complete */ wait_queue_head_t inocache_wq; + int inocache_hashsize; struct jffs2_inode_cache **inocache_list; spinlock_t inocache_lock; diff --git a/fs/jffs2/nodelist.c b/fs/jffs2/nodelist.c index af02bd1..5e03233 100644 --- a/fs/jffs2/nodelist.c +++ b/fs/jffs2/nodelist.c @@ -420,7 +420,7 @@ struct jffs2_inode_cache *jffs2_get_ino_cache(struct jffs2_sb_info *c, uint32_t { struct jffs2_inode_cache *ret; - ret = c->inocache_list[ino % INOCACHE_HASHSIZE]; + ret = c->inocache_list[ino % c->inocache_hashsize]; while (ret && ret->ino < ino) { ret = ret->next; } @@ -441,7 +441,7 @@ void jffs2_add_ino_cache (struct jffs2_sb_info *c, struct jffs2_inode_cache *new dbg_inocache("add %p (ino #%u)\n", new, new->ino); - prev = &c->inocache_list[new->ino % INOCACHE_HASHSIZE]; + prev = &c->inocache_list[new->ino % c->inocache_hashsize]; while ((*prev) && (*prev)->ino < new->ino) { prev = &(*prev)->next; @@ -462,7 +462,7 @@ void jffs2_del_ino_cache(struct jffs2_sb_info *c, struct jffs2_inode_cache *old) dbg_inocache("del %p (ino #%u)\n", old, old->ino); spin_lock(&c->inocache_lock); - prev = &c->inocache_list[old->ino % INOCACHE_HASHSIZE]; + prev = &c->inocache_list[old->ino % c->inocache_hashsize]; while ((*prev) && (*prev)->ino < old->ino) { prev = &(*prev)->next; @@ -487,7 +487,7 @@ void jffs2_free_ino_caches(struct jffs2_sb_info *c) int i; struct jffs2_inode_cache *this, *next; - for (i=0; i<INOCACHE_HASHSIZE; i++) { + for (i=0; i < c->inocache_hashsize; i++) { this = c->inocache_list[i]; while (this) { next = this->next; diff --git a/fs/jffs2/nodelist.h b/fs/jffs2/nodelist.h index 523a916..5a53d9b 100644 --- a/fs/jffs2/nodelist.h +++ b/fs/jffs2/nodelist.h @@ -199,7 +199,8 @@ struct jffs2_inode_cache { #define RAWNODE_CLASS_XATTR_DATUM 1 #define RAWNODE_CLASS_XATTR_REF 2 -#define INOCACHE_HASHSIZE 128 +#define INOCACHE_HASHSIZE_MIN 128 +#define INOCACHE_HASHSIZE_MAX 1024 #define write_ofs(c) ((c)->nextblock->offset + (c)->sector_size - (c)->nextblock->free_size) diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index 46f870d..b632ddd 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -20,7 +20,7 @@ #include "summary.h" #include "debug.h" -#define DEFAULT_EMPTY_SCAN_SIZE 1024 +#define DEFAULT_EMPTY_SCAN_SIZE 256 #define noisy_printk(noise, args...) do { \ if (*(noise)) { \ @@ -435,7 +435,7 @@ static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblo unsigned char *buf, uint32_t buf_size, struct jffs2_summary *s) { struct jffs2_unknown_node *node; struct jffs2_unknown_node crcnode; - uint32_t ofs, prevofs; + uint32_t ofs, prevofs, max_ofs; uint32_t hdr_crc, buf_ofs, buf_len; int err; int noise = 0; @@ -550,12 +550,12 @@ static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblo /* We temporarily use 'ofs' as a pointer into the buffer/jeb */ ofs = 0; - - /* Scan only 4KiB of 0xFF before declaring it's empty */ - while(ofs < EMPTY_SCAN_SIZE(c->sector_size) && *(uint32_t *)(&buf[ofs]) == 0xFFFFFFFF) + max_ofs = EMPTY_SCAN_SIZE(c->sector_size); + /* Scan only EMPTY_SCAN_SIZE of 0xFF before declaring it's empty */ + while(ofs < max_ofs && *(uint32_t *)(&buf[ofs]) == 0xFFFFFFFF) ofs += 4; - if (ofs == EMPTY_SCAN_SIZE(c->sector_size)) { + if (ofs == max_ofs) { #ifdef CONFIG_JFFS2_FS_WRITEBUFFER if (jffs2_cleanmarker_oob(c)) { /* scan oob, take care of cleanmarker */ diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index d1ae5df..c86041b 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -179,12 +179,11 @@ static int jffs2_fill_super(struct super_block *sb, void *data, int silent) return ret; } -static int jffs2_get_sb(struct file_system_type *fs_type, +static struct dentry *jffs2_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_mtd(fs_type, flags, dev_name, data, jffs2_fill_super, - mnt); + return mount_mtd(fs_type, flags, dev_name, data, jffs2_fill_super); } static void jffs2_put_super (struct super_block *sb) @@ -229,7 +228,7 @@ static void jffs2_kill_sb(struct super_block *sb) static struct file_system_type jffs2_fs_type = { .owner = THIS_MODULE, .name = "jffs2", - .get_sb = jffs2_get_sb, + .mount = jffs2_mount, .kill_sb = jffs2_kill_sb, }; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 68eee2b..0669fc1 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -583,11 +583,10 @@ static int jfs_unfreeze(struct super_block *sb) return 0; } -static int jfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *jfs_do_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, jfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, jfs_fill_super); } static int jfs_sync_fs(struct super_block *sb, int wait) @@ -770,7 +769,7 @@ static const struct export_operations jfs_export_operations = { static struct file_system_type jfs_fs_type = { .owner = THIS_MODULE, .name = "jfs", - .get_sb = jfs_get_sb, + .mount = jfs_do_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -201,9 +201,8 @@ static const struct super_operations simple_super_operations = { * Common helper for pseudo-filesystems (sockfs, pipefs, bdev - stuff that * will never be mountable) */ -int get_sb_pseudo(struct file_system_type *fs_type, char *name, - const struct super_operations *ops, unsigned long magic, - struct vfsmount *mnt) +struct dentry *mount_pseudo(struct file_system_type *fs_type, char *name, + const struct super_operations *ops, unsigned long magic) { struct super_block *s = sget(fs_type, NULL, set_anon_super, NULL); struct dentry *dentry; @@ -211,7 +210,7 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, struct qstr d_name = {.name = name, .len = strlen(name)}; if (IS_ERR(s)) - return PTR_ERR(s); + return ERR_CAST(s); s->s_flags = MS_NOUSER; s->s_maxbytes = MAX_LFS_FILESIZE; @@ -241,12 +240,11 @@ int get_sb_pseudo(struct file_system_type *fs_type, char *name, d_instantiate(dentry, root); s->s_root = dentry; s->s_flags |= MS_ACTIVE; - simple_set_mnt(mnt, s); - return 0; + return dget(s->s_root); Enomem: deactivate_locked_super(s); - return -ENOMEM; + return ERR_PTR(-ENOMEM); } int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) @@ -951,7 +949,7 @@ EXPORT_SYMBOL(dcache_dir_lseek); EXPORT_SYMBOL(dcache_dir_open); EXPORT_SYMBOL(dcache_readdir); EXPORT_SYMBOL(generic_read_dir); -EXPORT_SYMBOL(get_sb_pseudo); +EXPORT_SYMBOL(mount_pseudo); EXPORT_SYMBOL(simple_write_begin); EXPORT_SYMBOL(simple_write_end); EXPORT_SYMBOL(simple_dir_inode_operations); @@ -186,7 +186,7 @@ void locks_release_private(struct file_lock *fl) EXPORT_SYMBOL_GPL(locks_release_private); /* Free a lock which is not in use. */ -static void locks_free_lock(struct file_lock *fl) +void locks_free_lock(struct file_lock *fl) { BUG_ON(waitqueue_active(&fl->fl_wait)); BUG_ON(!list_empty(&fl->fl_block)); @@ -195,6 +195,7 @@ static void locks_free_lock(struct file_lock *fl) locks_release_private(fl); kmem_cache_free(filelock_cache, fl); } +EXPORT_SYMBOL(locks_free_lock); void locks_init_lock(struct file_lock *fl) { @@ -234,11 +235,8 @@ static void locks_copy_private(struct file_lock *new, struct file_lock *fl) fl->fl_ops->fl_copy_lock(new, fl); new->fl_ops = fl->fl_ops; } - if (fl->fl_lmops) { - if (fl->fl_lmops->fl_copy_lock) - fl->fl_lmops->fl_copy_lock(new, fl); + if (fl->fl_lmops) new->fl_lmops = fl->fl_lmops; - } } /* @@ -1371,20 +1369,22 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) struct inode *inode = dentry->d_inode; int error, rdlease_count = 0, wrlease_count = 0; + lease = *flp; + + error = -EACCES; if ((current_fsuid() != inode->i_uid) && !capable(CAP_LEASE)) - return -EACCES; + goto out; + error = -EINVAL; if (!S_ISREG(inode->i_mode)) - return -EINVAL; + goto out; error = security_file_lock(filp, arg); if (error) - return error; + goto out; time_out_leases(inode); BUG_ON(!(*flp)->fl_lmops->fl_break); - lease = *flp; - if (arg != F_UNLCK) { error = -EAGAIN; if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0)) @@ -1425,8 +1425,9 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) goto out; if (my_before != NULL) { - *flp = *my_before; error = lease->fl_lmops->fl_change(my_before, arg); + if (!error) + *flp = *my_before; goto out; } @@ -1441,7 +1442,6 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) return 0; out: - locks_free_lock(lease); return error; } EXPORT_SYMBOL(generic_setlease); @@ -1493,17 +1493,16 @@ int vfs_setlease(struct file *filp, long arg, struct file_lock **lease) } EXPORT_SYMBOL_GPL(vfs_setlease); -/** - * fcntl_setlease - sets a lease on an open file - * @fd: open file descriptor - * @filp: file pointer - * @arg: type of lease to obtain - * - * Call this fcntl to establish a lease on the file. - * Note that you also need to call %F_SETSIG to - * receive a signal when the lease is broken. - */ -int fcntl_setlease(unsigned int fd, struct file *filp, long arg) +static int do_fcntl_delete_lease(struct file *filp) +{ + struct file_lock fl, *flp = &fl; + + lease_init(filp, F_UNLCK, flp); + + return vfs_setlease(filp, F_UNLCK, &flp); +} + +static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) { struct file_lock *fl; struct fasync_struct *new; @@ -1521,8 +1520,11 @@ int fcntl_setlease(unsigned int fd, struct file *filp, long arg) } lock_flocks(); error = __vfs_setlease(filp, arg, &fl); - if (error || arg == F_UNLCK) - goto out_unlock; + if (error) { + unlock_flocks(); + locks_free_lock(fl); + goto out_free_fasync; + } /* * fasync_insert_entry() returns the old entry if any. @@ -1538,18 +1540,35 @@ int fcntl_setlease(unsigned int fd, struct file *filp, long arg) fl->fl_type = F_UNLCK | F_INPROGRESS; fl->fl_break_time = jiffies - 10; time_out_leases(inode); - goto out_unlock; + } else { + error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); } - - error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); -out_unlock: unlock_flocks(); + +out_free_fasync: if (new) fasync_free(new); return error; } /** + * fcntl_setlease - sets a lease on an open file + * @fd: open file descriptor + * @filp: file pointer + * @arg: type of lease to obtain + * + * Call this fcntl to establish a lease on the file. + * Note that you also need to call %F_SETSIG to + * receive a signal when the lease is broken. + */ +int fcntl_setlease(unsigned int fd, struct file *filp, long arg) +{ + if (arg == F_UNLCK) + return do_fcntl_delete_lease(filp); + return do_fcntl_add_lease(fd, filp, arg); +} + +/** * flock_lock_file_wait - Apply a FLOCK-style lock to a file * @filp: The file to apply the lock to * @fl: The lock to be applied diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c index 9bd2ce2..92ca6fb 100644 --- a/fs/logfs/dev_bdev.c +++ b/fs/logfs/dev_bdev.c @@ -298,9 +298,9 @@ static int bdev_write_sb(struct super_block *sb, struct page *page) return sync_request(page, bdev, WRITE); } -static void bdev_put_device(struct super_block *sb) +static void bdev_put_device(struct logfs_super *s) { - close_bdev_exclusive(logfs_super(sb)->s_bdev, FMODE_READ|FMODE_WRITE); + close_bdev_exclusive(s->s_bdev, FMODE_READ|FMODE_WRITE); } static int bdev_can_write_buf(struct super_block *sb, u64 ofs) @@ -320,8 +320,8 @@ static const struct logfs_device_ops bd_devops = { .put_device = bdev_put_device, }; -int logfs_get_sb_bdev(struct file_system_type *type, int flags, - const char *devname, struct vfsmount *mnt) +int logfs_get_sb_bdev(struct logfs_super *p, struct file_system_type *type, + const char *devname) { struct block_device *bdev; @@ -332,8 +332,11 @@ int logfs_get_sb_bdev(struct file_system_type *type, int flags, if (MAJOR(bdev->bd_dev) == MTD_BLOCK_MAJOR) { int mtdnr = MINOR(bdev->bd_dev); close_bdev_exclusive(bdev, FMODE_READ|FMODE_WRITE); - return logfs_get_sb_mtd(type, flags, mtdnr, mnt); + return logfs_get_sb_mtd(p, mtdnr); } - return logfs_get_sb_device(type, flags, NULL, bdev, &bd_devops, mnt); + p->s_bdev = bdev; + p->s_mtd = NULL; + p->s_devops = &bd_devops; + return 0; } diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c index a85d47d..7466e9d 100644 --- a/fs/logfs/dev_mtd.c +++ b/fs/logfs/dev_mtd.c @@ -230,9 +230,9 @@ static void mtd_writeseg(struct super_block *sb, u64 ofs, size_t len) __mtd_writeseg(sb, ofs, ofs >> PAGE_SHIFT, len >> PAGE_SHIFT); } -static void mtd_put_device(struct super_block *sb) +static void mtd_put_device(struct logfs_super *s) { - put_mtd_device(logfs_super(sb)->s_mtd); + put_mtd_device(s->s_mtd); } static int mtd_can_write_buf(struct super_block *sb, u64 ofs) @@ -265,14 +265,14 @@ static const struct logfs_device_ops mtd_devops = { .put_device = mtd_put_device, }; -int logfs_get_sb_mtd(struct file_system_type *type, int flags, - int mtdnr, struct vfsmount *mnt) +int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) { - struct mtd_info *mtd; - const struct logfs_device_ops *devops = &mtd_devops; - - mtd = get_mtd_device(NULL, mtdnr); + struct mtd_info *mtd = get_mtd_device(NULL, mtdnr); if (IS_ERR(mtd)) return PTR_ERR(mtd); - return logfs_get_sb_device(type, flags, mtd, NULL, devops, mnt); + + s->s_bdev = NULL; + s->s_mtd = mtd; + s->s_devops = &mtd_devops; + return 0; } diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index b878626..cd51a36 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -136,6 +136,7 @@ struct logfs_area_ops { int (*erase_segment)(struct logfs_area *area); }; +struct logfs_super; /* forward */ /** * struct logfs_device_ops - device access operations * @@ -156,7 +157,7 @@ struct logfs_device_ops { int ensure_write); int (*can_write_buf)(struct super_block *sb, u64 ofs); void (*sync)(struct super_block *sb); - void (*put_device)(struct super_block *sb); + void (*put_device)(struct logfs_super *s); }; /** @@ -471,11 +472,13 @@ void logfs_compr_exit(void); /* dev_bdev.c */ #ifdef CONFIG_BLOCK -int logfs_get_sb_bdev(struct file_system_type *type, int flags, - const char *devname, struct vfsmount *mnt); +int logfs_get_sb_bdev(struct logfs_super *s, + struct file_system_type *type, + const char *devname); #else -static inline int logfs_get_sb_bdev(struct file_system_type *type, int flags, - const char *devname, struct vfsmount *mnt) +static inline int logfs_get_sb_bdev(struct logfs_super *s, + struct file_system_type *type, + const char *devname) { return -ENODEV; } @@ -483,11 +486,9 @@ static inline int logfs_get_sb_bdev(struct file_system_type *type, int flags, /* dev_mtd.c */ #ifdef CONFIG_MTD -int logfs_get_sb_mtd(struct file_system_type *type, int flags, - int mtdnr, struct vfsmount *mnt); +int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) #else -static inline int logfs_get_sb_mtd(struct file_system_type *type, int flags, - int mtdnr, struct vfsmount *mnt) +static inline int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) { return -ENODEV; } @@ -619,9 +620,6 @@ void emergency_read_end(struct page *page); void logfs_crash_dump(struct super_block *sb); void *memchr_inv(const void *s, int c, size_t n); int logfs_statfs(struct dentry *dentry, struct kstatfs *stats); -int logfs_get_sb_device(struct file_system_type *type, int flags, - struct mtd_info *mtd, struct block_device *bdev, - const struct logfs_device_ops *devops, struct vfsmount *mnt); int logfs_check_ds(struct logfs_disk_super *ds); int logfs_write_sb(struct super_block *sb); diff --git a/fs/logfs/super.c b/fs/logfs/super.c index 5336155..33435e4 100644 --- a/fs/logfs/super.c +++ b/fs/logfs/super.c @@ -325,7 +325,7 @@ static int logfs_make_writeable(struct super_block *sb) return 0; } -static int logfs_get_sb_final(struct super_block *sb, struct vfsmount *mnt) +static int logfs_get_sb_final(struct super_block *sb) { struct logfs_super *super = logfs_super(sb); struct inode *rootdir; @@ -356,7 +356,6 @@ static int logfs_get_sb_final(struct super_block *sb, struct vfsmount *mnt) } log_super("LogFS: Finished mounting\n"); - simple_set_mnt(mnt, sb); return 0; fail: @@ -529,43 +528,37 @@ static void logfs_kill_sb(struct super_block *sb) logfs_cleanup_rw(sb); if (super->s_erase_page) __free_page(super->s_erase_page); - super->s_devops->put_device(sb); + super->s_devops->put_device(super); logfs_mempool_destroy(super->s_btree_pool); logfs_mempool_destroy(super->s_alias_pool); kfree(super); log_super("LogFS: Finished unmounting\n"); } -int logfs_get_sb_device(struct file_system_type *type, int flags, - struct mtd_info *mtd, struct block_device *bdev, - const struct logfs_device_ops *devops, struct vfsmount *mnt) +static struct dentry *logfs_get_sb_device(struct logfs_super *super, + struct file_system_type *type, int flags) { - struct logfs_super *super; struct super_block *sb; int err = -ENOMEM; static int mount_count; log_super("LogFS: Start mount %x\n", mount_count++); - super = kzalloc(sizeof(*super), GFP_KERNEL); - if (!super) - goto err0; - super->s_mtd = mtd; - super->s_bdev = bdev; err = -EINVAL; sb = sget(type, logfs_sb_test, logfs_sb_set, super); - if (IS_ERR(sb)) - goto err0; + if (IS_ERR(sb)) { + super->s_devops->put_device(super); + kfree(super); + return ERR_CAST(sb); + } if (sb->s_root) { /* Device is already in use */ - err = 0; - simple_set_mnt(mnt, sb); - goto err0; + super->s_devops->put_device(super); + kfree(super); + return dget(sb->s_root); } - super->s_devops = devops; - /* * sb->s_maxbytes is limited to 8TB. On 32bit systems, the page cache * only covers 16TB and the upper 8TB are used for indirect blocks. @@ -581,10 +574,12 @@ int logfs_get_sb_device(struct file_system_type *type, int flags, goto err1; sb->s_flags |= MS_ACTIVE; - err = logfs_get_sb_final(sb, mnt); - if (err) + err = logfs_get_sb_final(sb); + if (err) { deactivate_locked_super(sb); - return err; + return ERR_PTR(err); + } + return dget(sb->s_root); err1: /* no ->s_root, no ->put_super() */ @@ -592,37 +587,45 @@ err1: iput(super->s_segfile_inode); iput(super->s_mapping_inode); deactivate_locked_super(sb); - return err; -err0: - kfree(super); - //devops->put_device(sb); - return err; + return ERR_PTR(err); } -static int logfs_get_sb(struct file_system_type *type, int flags, - const char *devname, void *data, struct vfsmount *mnt) +static struct dentry *logfs_mount(struct file_system_type *type, int flags, + const char *devname, void *data) { ulong mtdnr; + struct logfs_super *super; + int err; - if (!devname) - return logfs_get_sb_bdev(type, flags, devname, mnt); - if (strncmp(devname, "mtd", 3)) - return logfs_get_sb_bdev(type, flags, devname, mnt); + super = kzalloc(sizeof(*super), GFP_KERNEL); + if (!super) + return ERR_PTR(-ENOMEM); - { + if (!devname) + err = logfs_get_sb_bdev(super, type, devname); + else if (strncmp(devname, "mtd", 3)) + err = logfs_get_sb_bdev(super, type, devname); + else { char *garbage; mtdnr = simple_strtoul(devname+3, &garbage, 0); if (*garbage) - return -EINVAL; + err = -EINVAL; + else + err = logfs_get_sb_mtd(super, mtdnr); + } + + if (err) { + kfree(super); + return ERR_PTR(err); } - return logfs_get_sb_mtd(type, flags, mtdnr, mnt); + return logfs_get_sb_device(super, type, flags); } static struct file_system_type logfs_fs_type = { .owner = THIS_MODULE, .name = "logfs", - .get_sb = logfs_get_sb, + .mount = logfs_mount, .kill_sb = logfs_kill_sb, .fs_flags = FS_REQUIRES_DEV, diff --git a/fs/minix/inode.c b/fs/minix/inode.c index e39d6bf..fb20208 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -614,17 +614,16 @@ void minix_truncate(struct inode * inode) V2_minix_truncate(inode); } -static int minix_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *minix_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, minix_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, minix_fill_super); } static struct file_system_type minix_fs_type = { .owner = THIS_MODULE, .name = "minix", - .get_sb = minix_get_sb, + .mount = minix_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -1574,6 +1574,7 @@ static struct file *finish_open(struct nameidata *nd, */ if (will_truncate) mnt_drop_write(nd->path.mnt); + path_put(&nd->path); return filp; exit: @@ -1675,6 +1676,7 @@ static struct file *do_last(struct nameidata *nd, struct path *path, } filp = nameidata_to_filp(nd); mnt_drop_write(nd->path.mnt); + path_put(&nd->path); if (!IS_ERR(filp)) { error = ima_file_check(filp, acc_mode); if (error) { diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 985fabb..d290545 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -1020,16 +1020,16 @@ out: return result; } -static int ncp_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ncp_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, ncp_fill_super, mnt); + return mount_nodev(fs_type, flags, data, ncp_fill_super); } static struct file_system_type ncp_fs_type = { .owner = THIS_MODULE, .name = "ncpfs", - .get_sb = ncp_get_sb, + .mount = ncp_mount, .kill_sb = kill_anon_super, .fs_flags = FS_BINARY_MOUNTDATA, }; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index e756075..60677f9 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -884,6 +884,5 @@ static int nfs_setlease(struct file *file, long arg, struct file_lock **fl) dprintk("NFS: setlease(%s/%s, arg=%ld)\n", file->f_path.dentry->d_parent->d_name.name, file->f_path.dentry->d_name.name, arg); - return -EINVAL; } diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 3600ec7..0a42e8f 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -260,8 +260,8 @@ static int nfs_statfs(struct dentry *, struct kstatfs *); static int nfs_show_options(struct seq_file *, struct vfsmount *); static int nfs_show_stats(struct seq_file *, struct vfsmount *); static int nfs_get_sb(struct file_system_type *, int, const char *, void *, struct vfsmount *); -static int nfs_xdev_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static struct dentry *nfs_xdev_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data); static void nfs_put_super(struct super_block *); static void nfs_kill_super(struct super_block *); static int nfs_remount(struct super_block *sb, int *flags, char *raw_data); @@ -277,7 +277,7 @@ static struct file_system_type nfs_fs_type = { struct file_system_type nfs_xdev_fs_type = { .owner = THIS_MODULE, .name = "nfs", - .get_sb = nfs_xdev_get_sb, + .mount = nfs_xdev_mount, .kill_sb = nfs_kill_super, .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, }; @@ -302,14 +302,14 @@ static int nfs4_try_mount(int flags, const char *dev_name, struct nfs_parsed_mount_data *data, struct vfsmount *mnt); static int nfs4_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); -static int nfs4_remote_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); -static int nfs4_xdev_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static struct dentry *nfs4_remote_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data); +static struct dentry *nfs4_xdev_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data); static int nfs4_referral_get_sb(struct file_system_type *fs_type, int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); -static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt); +static struct dentry *nfs4_remote_referral_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *raw_data); static void nfs4_kill_super(struct super_block *sb); static struct file_system_type nfs4_fs_type = { @@ -323,7 +323,7 @@ static struct file_system_type nfs4_fs_type = { static struct file_system_type nfs4_remote_fs_type = { .owner = THIS_MODULE, .name = "nfs4", - .get_sb = nfs4_remote_get_sb, + .mount = nfs4_remote_mount, .kill_sb = nfs4_kill_super, .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, }; @@ -331,7 +331,7 @@ static struct file_system_type nfs4_remote_fs_type = { struct file_system_type nfs4_xdev_fs_type = { .owner = THIS_MODULE, .name = "nfs4", - .get_sb = nfs4_xdev_get_sb, + .mount = nfs4_xdev_mount, .kill_sb = nfs4_kill_super, .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, }; @@ -339,7 +339,7 @@ struct file_system_type nfs4_xdev_fs_type = { static struct file_system_type nfs4_remote_referral_fs_type = { .owner = THIS_MODULE, .name = "nfs4", - .get_sb = nfs4_remote_referral_get_sb, + .mount = nfs4_remote_referral_mount, .kill_sb = nfs4_kill_super, .fs_flags = FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA, }; @@ -2397,9 +2397,9 @@ static void nfs_kill_super(struct super_block *s) /* * Clone an NFS2/3 server record on xdev traversal (FSID-change) */ -static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *raw_data, - struct vfsmount *mnt) +static struct dentry * +nfs_xdev_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data) { struct nfs_clone_mount *data = raw_data; struct super_block *s; @@ -2411,7 +2411,7 @@ static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, }; int error; - dprintk("--> nfs_xdev_get_sb()\n"); + dprintk("--> nfs_xdev_mount()\n"); /* create a new volume representation */ server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr); @@ -2458,28 +2458,26 @@ static int nfs_xdev_get_sb(struct file_system_type *fs_type, int flags, } s->s_flags |= MS_ACTIVE; - mnt->mnt_sb = s; - mnt->mnt_root = mntroot; /* clone any lsm security options from the parent to the new sb */ security_sb_clone_mnt_opts(data->sb, s); - dprintk("<-- nfs_xdev_get_sb() = 0\n"); - return 0; + dprintk("<-- nfs_xdev_mount() = 0\n"); + return mntroot; out_err_nosb: nfs_free_server(server); out_err_noserver: - dprintk("<-- nfs_xdev_get_sb() = %d [error]\n", error); - return error; + dprintk("<-- nfs_xdev_mount() = %d [error]\n", error); + return ERR_PTR(error); error_splat_super: if (server && !s->s_root) bdi_unregister(&server->backing_dev_info); error_splat_bdi: deactivate_locked_super(s); - dprintk("<-- nfs_xdev_get_sb() = %d [splat]\n", error); - return error; + dprintk("<-- nfs_xdev_mount() = %d [splat]\n", error); + return ERR_PTR(error); } #ifdef CONFIG_NFS_V4 @@ -2649,8 +2647,9 @@ out_no_address: /* * Get the superblock for the NFS4 root partition */ -static int nfs4_remote_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt) +static struct dentry * +nfs4_remote_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data) { struct nfs_parsed_mount_data *data = raw_data; struct super_block *s; @@ -2714,15 +2713,16 @@ static int nfs4_remote_get_sb(struct file_system_type *fs_type, goto error_splat_root; s->s_flags |= MS_ACTIVE; - mnt->mnt_sb = s; - mnt->mnt_root = mntroot; - error = 0; + + security_free_mnt_opts(&data->lsm_opts); + nfs_free_fhandle(mntfh); + return mntroot; out: security_free_mnt_opts(&data->lsm_opts); out_free_fh: nfs_free_fhandle(mntfh); - return error; + return ERR_PTR(error); out_free: nfs_free_server(server); @@ -2968,9 +2968,9 @@ static void nfs4_kill_super(struct super_block *sb) /* * Clone an NFS4 server record on xdev traversal (FSID-change) */ -static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *raw_data, - struct vfsmount *mnt) +static struct dentry * +nfs4_xdev_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data) { struct nfs_clone_mount *data = raw_data; struct super_block *s; @@ -2982,7 +2982,7 @@ static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags, }; int error; - dprintk("--> nfs4_xdev_get_sb()\n"); + dprintk("--> nfs4_xdev_mount()\n"); /* create a new volume representation */ server = nfs_clone_server(NFS_SB(data->sb), data->fh, data->fattr); @@ -3029,32 +3029,30 @@ static int nfs4_xdev_get_sb(struct file_system_type *fs_type, int flags, } s->s_flags |= MS_ACTIVE; - mnt->mnt_sb = s; - mnt->mnt_root = mntroot; security_sb_clone_mnt_opts(data->sb, s); - dprintk("<-- nfs4_xdev_get_sb() = 0\n"); - return 0; + dprintk("<-- nfs4_xdev_mount() = 0\n"); + return mntroot; out_err_nosb: nfs_free_server(server); out_err_noserver: - dprintk("<-- nfs4_xdev_get_sb() = %d [error]\n", error); - return error; + dprintk("<-- nfs4_xdev_mount() = %d [error]\n", error); + return ERR_PTR(error); error_splat_super: if (server && !s->s_root) bdi_unregister(&server->backing_dev_info); error_splat_bdi: deactivate_locked_super(s); - dprintk("<-- nfs4_xdev_get_sb() = %d [splat]\n", error); - return error; + dprintk("<-- nfs4_xdev_mount() = %d [splat]\n", error); + return ERR_PTR(error); } -static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *raw_data, - struct vfsmount *mnt) +static struct dentry * +nfs4_remote_referral_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *raw_data) { struct nfs_clone_mount *data = raw_data; struct super_block *s; @@ -3118,14 +3116,12 @@ static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type, } s->s_flags |= MS_ACTIVE; - mnt->mnt_sb = s; - mnt->mnt_root = mntroot; security_sb_clone_mnt_opts(data->sb, s); nfs_free_fhandle(mntfh); dprintk("<-- nfs4_referral_get_sb() = 0\n"); - return 0; + return mntroot; out_err_nosb: nfs_free_server(server); @@ -3133,7 +3129,7 @@ out_err_noserver: nfs_free_fhandle(mntfh); out_err_nofh: dprintk("<-- nfs4_referral_get_sb() = %d [error]\n", error); - return error; + return ERR_PTR(error); error_splat_super: if (server && !s->s_root) @@ -3142,7 +3138,7 @@ error_splat_bdi: deactivate_locked_super(s); nfs_free_fhandle(mntfh); dprintk("<-- nfs4_referral_get_sb() = %d [splat]\n", error); - return error; + return ERR_PTR(error); } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 9a16bad..7bdec85 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -444,9 +444,9 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir, /* set up nfs_renamedata */ data->old_dir = old_dir; - atomic_inc(&old_dir->i_count); + ihold(old_dir); data->new_dir = new_dir; - atomic_inc(&new_dir->i_count); + ihold(new_dir); data->old_dentry = dget(old_dentry); data->new_dentry = dget(new_dentry); nfs_fattr_init(&data->old_fattr); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 56347e0..f1e5ec6 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2310,22 +2310,6 @@ void nfsd_release_deleg_cb(struct file_lock *fl) } /* - * Set the delegation file_lock back pointer. - * - * Called from setlease() with lock_kernel() held. - */ -static -void nfsd_copy_lock_deleg_cb(struct file_lock *new, struct file_lock *fl) -{ - struct nfs4_delegation *dp = (struct nfs4_delegation *)new->fl_owner; - - dprintk("NFSD: nfsd_copy_lock_deleg_cb: new fl %p dp %p\n", new, dp); - if (!dp) - return; - dp->dl_flock = new; -} - -/* * Called from setlease() with lock_kernel() held */ static @@ -2355,7 +2339,6 @@ int nfsd_change_deleg_cb(struct file_lock **onlist, int arg) static const struct lock_manager_operations nfsd_lease_mng_ops = { .fl_break = nfsd_break_deleg_cb, .fl_release_private = nfsd_release_deleg_cb, - .fl_copy_lock = nfsd_copy_lock_deleg_cb, .fl_mylease = nfsd_same_client_deleg_cb, .fl_change = nfsd_change_deleg_cb, }; @@ -2661,12 +2644,15 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta fl->fl_file = find_readable_file(stp->st_file); BUG_ON(!fl->fl_file); fl->fl_pid = current->tgid; + dp->dl_flock = fl; /* vfs_setlease checks to see if delegation should be handed out. * the lock_manager callbacks fl_mylease and fl_change are used */ if ((status = vfs_setlease(fl->fl_file, fl->fl_type, &fl))) { dprintk("NFSD: setlease failed [%d], no delegation\n", status); + dp->dl_flock = NULL; + locks_free_lock(fl); unhash_delegation(dp); flag = NFS4_OPEN_DELEGATE_NONE; goto out; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index d6dc3f6..4514ebb 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -1405,16 +1405,16 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent) return simple_fill_super(sb, 0x6e667364, nfsd_files); } -static int nfsd_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *nfsd_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_single(fs_type, flags, data, nfsd_fill_super, mnt); + return mount_single(fs_type, flags, data, nfsd_fill_super); } static struct file_system_type nfsd_fs_type = { .owner = THIS_MODULE, .name = "nfsd", - .get_sb = nfsd_get_sb, + .mount = nfsd_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 35ae03c..f804d41 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -1141,9 +1141,9 @@ static int nilfs_test_bdev_super(struct super_block *s, void *data) return (void *)s->s_bdev == data; } -static int -nilfs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry * +nilfs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { struct nilfs_super_data sd; struct super_block *s; @@ -1156,7 +1156,7 @@ nilfs_get_sb(struct file_system_type *fs_type, int flags, sd.bdev = open_bdev_exclusive(dev_name, mode, fs_type); if (IS_ERR(sd.bdev)) - return PTR_ERR(sd.bdev); + return ERR_CAST(sd.bdev); sd.cno = 0; sd.flags = flags; @@ -1235,9 +1235,7 @@ nilfs_get_sb(struct file_system_type *fs_type, int flags, if (!s_new) close_bdev_exclusive(sd.bdev, mode); - mnt->mnt_sb = s; - mnt->mnt_root = root_dentry; - return 0; + return root_dentry; failed_super: deactivate_locked_super(s); @@ -1245,13 +1243,13 @@ nilfs_get_sb(struct file_system_type *fs_type, int flags, failed: if (!s_new) close_bdev_exclusive(sd.bdev, mode); - return err; + return ERR_PTR(err); } struct file_system_type nilfs_fs_type = { .owner = THIS_MODULE, .name = "nilfs2", - .get_sb = nilfs_get_sb, + .mount = nilfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/notify/Kconfig b/fs/notify/Kconfig index b388443..22c629e 100644 --- a/fs/notify/Kconfig +++ b/fs/notify/Kconfig @@ -3,4 +3,4 @@ config FSNOTIFY source "fs/notify/dnotify/Kconfig" source "fs/notify/inotify/Kconfig" -#source "fs/notify/fanotify/Kconfig" +source "fs/notify/fanotify/Kconfig" diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c index 85366c7..b04f88e 100644 --- a/fs/notify/fanotify/fanotify.c +++ b/fs/notify/fanotify/fanotify.c @@ -131,6 +131,7 @@ static int fanotify_handle_event(struct fsnotify_group *group, BUILD_BUG_ON(FAN_Q_OVERFLOW != FS_Q_OVERFLOW); BUILD_BUG_ON(FAN_OPEN_PERM != FS_OPEN_PERM); BUILD_BUG_ON(FAN_ACCESS_PERM != FS_ACCESS_PERM); + BUILD_BUG_ON(FAN_ONDIR != FS_ISDIR); pr_debug("%s: group=%p event=%p\n", __func__, group, event); @@ -160,20 +161,21 @@ static bool fanotify_should_send_event(struct fsnotify_group *group, __u32 event_mask, void *data, int data_type) { __u32 marks_mask, marks_ignored_mask; + struct path *path = data; pr_debug("%s: group=%p to_tell=%p inode_mark=%p vfsmnt_mark=%p " "mask=%x data=%p data_type=%d\n", __func__, group, to_tell, inode_mark, vfsmnt_mark, event_mask, data, data_type); - /* sorry, fanotify only gives a damn about files and dirs */ - if (!S_ISREG(to_tell->i_mode) && - !S_ISDIR(to_tell->i_mode)) - return false; - /* if we don't have enough info to send an event to userspace say no */ if (data_type != FSNOTIFY_EVENT_PATH) return false; + /* sorry, fanotify only gives a damn about files and dirs */ + if (!S_ISREG(path->dentry->d_inode->i_mode) && + !S_ISDIR(path->dentry->d_inode->i_mode)) + return false; + if (inode_mark && vfsmnt_mark) { marks_mask = (vfsmnt_mark->mask | inode_mark->mask); marks_ignored_mask = (vfsmnt_mark->ignored_mask | inode_mark->ignored_mask); @@ -194,16 +196,29 @@ static bool fanotify_should_send_event(struct fsnotify_group *group, BUG(); } + if (S_ISDIR(path->dentry->d_inode->i_mode) && + (marks_ignored_mask & FS_ISDIR)) + return false; + if (event_mask & marks_mask & ~marks_ignored_mask) return true; return false; } +static void fanotify_free_group_priv(struct fsnotify_group *group) +{ + struct user_struct *user; + + user = group->fanotify_data.user; + atomic_dec(&user->fanotify_listeners); + free_uid(user); +} + const struct fsnotify_ops fanotify_fsnotify_ops = { .handle_event = fanotify_handle_event, .should_send_event = fanotify_should_send_event, - .free_group_priv = NULL, + .free_group_priv = fanotify_free_group_priv, .free_event_priv = NULL, .freeing_mark = NULL, }; diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index bbcb98e..0632248 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -16,6 +16,10 @@ #include <asm/ioctls.h> +#define FANOTIFY_DEFAULT_MAX_EVENTS 16384 +#define FANOTIFY_DEFAULT_MAX_MARKS 8192 +#define FANOTIFY_DEFAULT_MAX_LISTENERS 128 + extern const struct fsnotify_ops fanotify_fsnotify_ops; static struct kmem_cache *fanotify_mark_cache __read_mostly; @@ -326,7 +330,7 @@ static ssize_t fanotify_read(struct file *file, char __user *buf, ret = -EAGAIN; if (file->f_flags & O_NONBLOCK) break; - ret = -EINTR; + ret = -ERESTARTSYS; if (signal_pending(current)) break; @@ -372,11 +376,10 @@ static ssize_t fanotify_write(struct file *file, const char __user *buf, size_t static int fanotify_release(struct inode *ignored, struct file *file) { struct fsnotify_group *group = file->private_data; - struct fanotify_response_event *re, *lre; - - pr_debug("%s: file=%p group=%p\n", __func__, file, group); #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS + struct fanotify_response_event *re, *lre; + mutex_lock(&group->fanotify_data.access_mutex); group->fanotify_data.bypass_perm = true; @@ -554,18 +557,24 @@ static __u32 fanotify_mark_add_to_mask(struct fsnotify_mark *fsn_mark, __u32 mask, unsigned int flags) { - __u32 oldmask; + __u32 oldmask = -1; spin_lock(&fsn_mark->lock); if (!(flags & FAN_MARK_IGNORED_MASK)) { oldmask = fsn_mark->mask; fsnotify_set_mark_mask_locked(fsn_mark, (oldmask | mask)); } else { - oldmask = fsn_mark->ignored_mask; - fsnotify_set_mark_ignored_mask_locked(fsn_mark, (oldmask | mask)); + __u32 tmask = fsn_mark->ignored_mask | mask; + fsnotify_set_mark_ignored_mask_locked(fsn_mark, tmask); if (flags & FAN_MARK_IGNORED_SURV_MODIFY) fsn_mark->flags |= FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY; } + + if (!(flags & FAN_MARK_ONDIR)) { + __u32 tmask = fsn_mark->ignored_mask | FAN_ONDIR; + fsnotify_set_mark_ignored_mask_locked(fsn_mark, tmask); + } + spin_unlock(&fsn_mark->lock); return mask & ~oldmask; @@ -582,6 +591,9 @@ static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, if (!fsn_mark) { int ret; + if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks) + return -ENOSPC; + fsn_mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL); if (!fsn_mark) return -ENOMEM; @@ -610,10 +622,23 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group, pr_debug("%s: group=%p inode=%p\n", __func__, group, inode); + /* + * If some other task has this inode open for write we should not add + * an ignored mark, unless that ignored mark is supposed to survive + * modification changes anyway. + */ + if ((flags & FAN_MARK_IGNORED_MASK) && + !(flags & FAN_MARK_IGNORED_SURV_MODIFY) && + (atomic_read(&inode->i_writecount) > 0)) + return 0; + fsn_mark = fsnotify_find_inode_mark(group, inode); if (!fsn_mark) { int ret; + if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks) + return -ENOSPC; + fsn_mark = kmem_cache_alloc(fanotify_mark_cache, GFP_KERNEL); if (!fsn_mark) return -ENOMEM; @@ -637,6 +662,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) { struct fsnotify_group *group; int f_flags, fd; + struct user_struct *user; pr_debug("%s: flags=%d event_f_flags=%d\n", __func__, flags, event_f_flags); @@ -647,6 +673,12 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) if (flags & ~FAN_ALL_INIT_FLAGS) return -EINVAL; + user = get_current_user(); + if (atomic_read(&user->fanotify_listeners) > FANOTIFY_DEFAULT_MAX_LISTENERS) { + free_uid(user); + return -EMFILE; + } + f_flags = O_RDWR | FMODE_NONOTIFY; if (flags & FAN_CLOEXEC) f_flags |= O_CLOEXEC; @@ -658,12 +690,47 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) if (IS_ERR(group)) return PTR_ERR(group); + group->fanotify_data.user = user; + atomic_inc(&user->fanotify_listeners); + group->fanotify_data.f_flags = event_f_flags; #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS mutex_init(&group->fanotify_data.access_mutex); init_waitqueue_head(&group->fanotify_data.access_waitq); INIT_LIST_HEAD(&group->fanotify_data.access_list); #endif + switch (flags & FAN_ALL_CLASS_BITS) { + case FAN_CLASS_NOTIF: + group->priority = FS_PRIO_0; + break; + case FAN_CLASS_CONTENT: + group->priority = FS_PRIO_1; + break; + case FAN_CLASS_PRE_CONTENT: + group->priority = FS_PRIO_2; + break; + default: + fd = -EINVAL; + goto out_put_group; + } + + if (flags & FAN_UNLIMITED_QUEUE) { + fd = -EPERM; + if (!capable(CAP_SYS_ADMIN)) + goto out_put_group; + group->max_events = UINT_MAX; + } else { + group->max_events = FANOTIFY_DEFAULT_MAX_EVENTS; + } + + if (flags & FAN_UNLIMITED_MARKS) { + fd = -EPERM; + if (!capable(CAP_SYS_ADMIN)) + goto out_put_group; + group->fanotify_data.max_marks = UINT_MAX; + } else { + group->fanotify_data.max_marks = FANOTIFY_DEFAULT_MAX_MARKS; + } fd = anon_inode_getfd("[fanotify]", &fanotify_fops, group, f_flags); if (fd < 0) @@ -704,6 +771,12 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, default: return -EINVAL; } + + if (mask & FAN_ONDIR) { + flags |= FAN_MARK_ONDIR; + mask &= ~FAN_ONDIR; + } + #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS if (mask & ~(FAN_ALL_EVENTS | FAN_ALL_PERM_EVENTS | FAN_EVENT_ON_CHILD)) #else @@ -719,6 +792,16 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, ret = -EINVAL; if (unlikely(filp->f_op != &fanotify_fops)) goto fput_and_out; + group = filp->private_data; + + /* + * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF. These are not + * allowed to set permissions events. + */ + ret = -EINVAL; + if (mask & FAN_ALL_PERM_EVENTS && + group->priority == FS_PRIO_0) + goto fput_and_out; ret = fanotify_find_path(dfd, pathname, &path, flags); if (ret) @@ -729,7 +812,6 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, inode = path.dentry->d_inode; else mnt = path.mnt; - group = filp->private_data; /* create/update an inode mark */ switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) { diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 4498a20..20dc218 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -84,16 +84,17 @@ void __fsnotify_update_child_dentry_flags(struct inode *inode) } /* Notify this dentry's parent about a child's events. */ -void __fsnotify_parent(struct path *path, struct dentry *dentry, __u32 mask) +int __fsnotify_parent(struct path *path, struct dentry *dentry, __u32 mask) { struct dentry *parent; struct inode *p_inode; + int ret = 0; if (!dentry) dentry = path->dentry; if (!(dentry->d_flags & DCACHE_FSNOTIFY_PARENT_WATCHED)) - return; + return 0; parent = dget_parent(dentry); p_inode = parent->d_inode; @@ -106,14 +107,16 @@ void __fsnotify_parent(struct path *path, struct dentry *dentry, __u32 mask) mask |= FS_EVENT_ON_CHILD; if (path) - fsnotify(p_inode, mask, path, FSNOTIFY_EVENT_PATH, - dentry->d_name.name, 0); + ret = fsnotify(p_inode, mask, path, FSNOTIFY_EVENT_PATH, + dentry->d_name.name, 0); else - fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE, - dentry->d_name.name, 0); + ret = fsnotify(p_inode, mask, dentry->d_inode, FSNOTIFY_EVENT_INODE, + dentry->d_name.name, 0); } dput(parent); + + return ret; } EXPORT_SYMBOL_GPL(__fsnotify_parent); @@ -252,20 +255,23 @@ int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is, if (inode_group > vfsmount_group) { /* handle inode */ - send_to_group(to_tell, NULL, inode_mark, NULL, mask, data, - data_is, cookie, file_name, &event); + ret = send_to_group(to_tell, NULL, inode_mark, NULL, mask, data, + data_is, cookie, file_name, &event); /* we didn't use the vfsmount_mark */ vfsmount_group = NULL; } else if (vfsmount_group > inode_group) { - send_to_group(to_tell, mnt, NULL, vfsmount_mark, mask, data, - data_is, cookie, file_name, &event); + ret = send_to_group(to_tell, mnt, NULL, vfsmount_mark, mask, data, + data_is, cookie, file_name, &event); inode_group = NULL; } else { - send_to_group(to_tell, mnt, inode_mark, vfsmount_mark, - mask, data, data_is, cookie, file_name, - &event); + ret = send_to_group(to_tell, mnt, inode_mark, vfsmount_mark, + mask, data, data_is, cookie, file_name, + &event); } + if (ret && (mask & ALL_FSNOTIFY_PERM_EVENTS)) + goto out; + if (inode_group) inode_node = srcu_dereference(inode_node->next, &fsnotify_mark_srcu); @@ -273,7 +279,8 @@ int fsnotify(struct inode *to_tell, __u32 mask, void *data, int data_is, vfsmount_node = srcu_dereference(vfsmount_node->next, &fsnotify_mark_srcu); } - + ret = 0; +out: srcu_read_unlock(&fsnotify_mark_srcu, idx); /* * fsnotify_create_event() took a reference so the event can't be cleaned diff --git a/fs/notify/inode_mark.c b/fs/notify/inode_mark.c index 21ed106..4c29fcf 100644 --- a/fs/notify/inode_mark.c +++ b/fs/notify/inode_mark.c @@ -177,7 +177,8 @@ void fsnotify_set_inode_mark_mask_locked(struct fsnotify_mark *mark, * Attach an initialized mark to a given inode. * These marks may be used for the fsnotify backend to determine which * event types should be delivered to which group and for which inodes. These - * marks are ordered according to the group's location in memory. + * marks are ordered according to priority, highest number first, and then by + * the group's location in memory. */ int fsnotify_add_inode_mark(struct fsnotify_mark *mark, struct fsnotify_group *group, struct inode *inode, @@ -211,7 +212,11 @@ int fsnotify_add_inode_mark(struct fsnotify_mark *mark, goto out; } - if (mark->group < lmark->group) + if (mark->group->priority < lmark->group->priority) + continue; + + if ((mark->group->priority == lmark->group->priority) && + (mark->group < lmark->group)) continue; hlist_add_before_rcu(&mark->i.i_list, &lmark->i.i_list); diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 24edc11..444c305 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -862,7 +862,7 @@ static int __init inotify_user_setup(void) BUILD_BUG_ON(IN_Q_OVERFLOW != FS_Q_OVERFLOW); BUILD_BUG_ON(IN_IGNORED != FS_IN_IGNORED); BUILD_BUG_ON(IN_EXCL_UNLINK != FS_EXCL_UNLINK); - BUILD_BUG_ON(IN_ISDIR != FS_IN_ISDIR); + BUILD_BUG_ON(IN_ISDIR != FS_ISDIR); BUILD_BUG_ON(IN_ONESHOT != FS_IN_ONESHOT); BUG_ON(hweight32(ALL_INOTIFY_BITS) != 21); diff --git a/fs/notify/vfsmount_mark.c b/fs/notify/vfsmount_mark.c index 56772b5..85eebff 100644 --- a/fs/notify/vfsmount_mark.c +++ b/fs/notify/vfsmount_mark.c @@ -169,7 +169,11 @@ int fsnotify_add_vfsmount_mark(struct fsnotify_mark *mark, goto out; } - if (mark->group < lmark->group) + if (mark->group->priority < lmark->group->priority) + continue; + + if ((mark->group->priority == lmark->group->priority) && + (mark->group < lmark->group)) continue; hlist_add_before_rcu(&mark->m.m_list, &lmark->m.m_list); diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index d3fbe57..a30ecacc 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -3059,17 +3059,16 @@ struct kmem_cache *ntfs_index_ctx_cache; /* Driver wide mutex. */ DEFINE_MUTEX(ntfs_lock); -static int ntfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ntfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ntfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, ntfs_fill_super); } static struct file_system_type ntfs_fs_type = { .owner = THIS_MODULE, .name = "ntfs", - .get_sb = ntfs_get_sb, + .mount = ntfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index 75e115f..b2df490 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -643,16 +643,16 @@ static const struct inode_operations dlmfs_file_inode_operations = { .setattr = dlmfs_file_setattr, }; -static int dlmfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *dlmfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, dlmfs_fill_super, mnt); + return mount_nodev(fs_type, flags, data, dlmfs_fill_super); } static struct file_system_type dlmfs_fs_type = { .owner = THIS_MODULE, .name = "ocfs2_dlmfs", - .get_sb = dlmfs_get_sb, + .mount = dlmfs_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 56f0cb3..f02c0ef 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1236,14 +1236,12 @@ read_super_error: return status; } -static int ocfs2_get_sb(struct file_system_type *fs_type, +static struct dentry *ocfs2_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, - struct vfsmount *mnt) + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ocfs2_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, ocfs2_fill_super); } static void ocfs2_kill_sb(struct super_block *sb) @@ -1267,8 +1265,7 @@ out: static struct file_system_type ocfs2_fs_type = { .owner = THIS_MODULE, .name = "ocfs2", - .get_sb = ocfs2_get_sb, /* is this called when we mount - * the fs? */ + .mount = ocfs2_mount, .kill_sb = ocfs2_kill_sb, .fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE, diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c index 14a2286..e043c4c 100644 --- a/fs/omfs/inode.c +++ b/fs/omfs/inode.c @@ -557,17 +557,16 @@ end: return ret; } -static int omfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data, struct vfsmount *m) +static struct dentry *omfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, omfs_fill_super, m); + return mount_bdev(fs_type, flags, dev_name, data, omfs_fill_super); } static struct file_system_type omfs_fs_type = { .owner = THIS_MODULE, .name = "omfs", - .get_sb = omfs_get_sb, + .mount = omfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -786,11 +786,11 @@ struct file *nameidata_to_filp(struct nameidata *nd) /* Pick up the filp from the open intent */ filp = nd->intent.open.file; /* Has the filesystem initialised the file for us? */ - if (filp->f_path.dentry == NULL) + if (filp->f_path.dentry == NULL) { + path_get(&nd->path); filp = __dentry_open(nd->path.dentry, nd->path.mnt, filp, NULL, cred); - else - path_put(&nd->path); + } return filp; } diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index ffcd04f..ddb1f41 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -415,16 +415,16 @@ out_no_root: return ret; } -static int openprom_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *openprom_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_single(fs_type, flags, data, openprom_fill_super, mnt); + return mount_single(fs_type, flags, data, openprom_fill_super) } static struct file_system_type openprom_fs_type = { .owner = THIS_MODULE, .name = "openpromfs", - .get_sb = openprom_get_sb, + .mount = openprom_mount, .kill_sb = kill_anon_super, }; @@ -1247,16 +1247,15 @@ out: * any operations on the root directory. However, we need a non-trivial * d_name - pipe: will go nicely and kill the special-casing in procfs. */ -static int pipefs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *pipefs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_pseudo(fs_type, "pipe:", NULL, PIPEFS_MAGIC, mnt); + return mount_pseudo(fs_type, "pipe:", NULL, PIPEFS_MAGIC); } static struct file_system_type pipe_fs_type = { .name = "pipefs", - .get_sb = pipefs_get_sb, + .mount = pipefs_mount, .kill_sb = kill_anon_super, }; diff --git a/fs/proc/root.c b/fs/proc/root.c index 93d99b3..ef9fa8e 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -35,8 +35,8 @@ static int proc_set_super(struct super_block *sb, void *data) return set_anon_super(sb, NULL); } -static int proc_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *proc_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { int err; struct super_block *sb; @@ -61,14 +61,14 @@ static int proc_get_sb(struct file_system_type *fs_type, sb = sget(fs_type, proc_test_super, proc_set_super, ns); if (IS_ERR(sb)) - return PTR_ERR(sb); + return ERR_CAST(sb); if (!sb->s_root) { sb->s_flags = flags; err = proc_fill_super(sb); if (err) { deactivate_locked_super(sb); - return err; + return ERR_PTR(err); } ei = PROC_I(sb->s_root->d_inode); @@ -79,11 +79,9 @@ static int proc_get_sb(struct file_system_type *fs_type, } sb->s_flags |= MS_ACTIVE; - ns->proc_mnt = mnt; } - simple_set_mnt(mnt, sb); - return 0; + return dget(sb->s_root); } static void proc_kill_sb(struct super_block *sb) @@ -97,7 +95,7 @@ static void proc_kill_sb(struct super_block *sb) static struct file_system_type proc_fs_type = { .name = "proc", - .get_sb = proc_get_sb, + .mount = proc_mount, .kill_sb = proc_kill_sb, }; @@ -115,6 +113,7 @@ void __init proc_root_init(void) return; } + init_pid_ns.proc_mnt = proc_mnt; proc_symlink("mounts", NULL, "self/mounts"); proc_net_init(); @@ -213,6 +212,7 @@ int pid_ns_prepare_proc(struct pid_namespace *ns) if (IS_ERR(mnt)) return PTR_ERR(mnt); + ns->proc_mnt = mnt; return 0; } diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 01bad30..fcada42 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -454,17 +454,16 @@ static void destroy_inodecache(void) kmem_cache_destroy(qnx4_inode_cachep); } -static int qnx4_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *qnx4_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, qnx4_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, qnx4_fill_super); } static struct file_system_type qnx4_fs_type = { .owner = THIS_MODULE, .name = "qnx4", - .get_sb = qnx4_get_sb, + .mount = qnx4_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index 67fadb1..eacb166 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -255,17 +255,16 @@ fail: return err; } -int ramfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +struct dentry *ramfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags, data, ramfs_fill_super, mnt); + return mount_nodev(fs_type, flags, data, ramfs_fill_super); } -static int rootfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *rootfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_nodev(fs_type, flags|MS_NOUSER, data, ramfs_fill_super, - mnt); + return mount_nodev(fs_type, flags|MS_NOUSER, data, ramfs_fill_super); } static void ramfs_kill_sb(struct super_block *sb) @@ -276,12 +275,12 @@ static void ramfs_kill_sb(struct super_block *sb) static struct file_system_type ramfs_fs_type = { .name = "ramfs", - .get_sb = ramfs_get_sb, + .mount = ramfs_mount, .kill_sb = ramfs_kill_sb, }; static struct file_system_type rootfs_fs_type = { .name = "rootfs", - .get_sb = rootfs_get_sb, + .mount = rootfs_mount, .kill_sb = kill_litter_super, }; diff --git a/fs/read_write.c b/fs/read_write.c index 9cd9d14..431a0ed 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -243,8 +243,6 @@ bad: * them to something that fits in "int" so that others * won't have to do range checks all the time. */ -#define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK) - int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count) { struct inode *inode; @@ -584,65 +582,71 @@ ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, unsigned long nr_segs, unsigned long fast_segs, struct iovec *fast_pointer, struct iovec **ret_pointer) - { +{ unsigned long seg; - ssize_t ret; + ssize_t ret; struct iovec *iov = fast_pointer; - /* - * SuS says "The readv() function *may* fail if the iovcnt argument - * was less than or equal to 0, or greater than {IOV_MAX}. Linux has - * traditionally returned zero for zero segments, so... - */ + /* + * SuS says "The readv() function *may* fail if the iovcnt argument + * was less than or equal to 0, or greater than {IOV_MAX}. Linux has + * traditionally returned zero for zero segments, so... + */ if (nr_segs == 0) { ret = 0; - goto out; + goto out; } - /* - * First get the "struct iovec" from user memory and - * verify all the pointers - */ + /* + * First get the "struct iovec" from user memory and + * verify all the pointers + */ if (nr_segs > UIO_MAXIOV) { ret = -EINVAL; - goto out; + goto out; } if (nr_segs > fast_segs) { - iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL); + iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL); if (iov == NULL) { ret = -ENOMEM; - goto out; + goto out; } - } + } if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector))) { ret = -EFAULT; - goto out; + goto out; } - /* + /* * According to the Single Unix Specification we should return EINVAL * if an element length is < 0 when cast to ssize_t or if the * total length would overflow the ssize_t return value of the * system call. - */ + * + * Linux caps all read/write calls to MAX_RW_COUNT, and avoids the + * overflow case. + */ ret = 0; - for (seg = 0; seg < nr_segs; seg++) { - void __user *buf = iov[seg].iov_base; - ssize_t len = (ssize_t)iov[seg].iov_len; + for (seg = 0; seg < nr_segs; seg++) { + void __user *buf = iov[seg].iov_base; + ssize_t len = (ssize_t)iov[seg].iov_len; /* see if we we're about to use an invalid len or if * it's about to overflow ssize_t */ - if (len < 0 || (ret + len < ret)) { + if (len < 0) { ret = -EINVAL; - goto out; + goto out; } if (unlikely(!access_ok(vrfy_dir(type), buf, len))) { ret = -EFAULT; - goto out; + goto out; + } + if (len > MAX_RW_COUNT - ret) { + len = MAX_RW_COUNT - ret; + iov[seg].iov_len = len; } - ret += len; - } + } out: *ret_pointer = iov; return ret; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index e15ff612..3bf7a64 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -2213,12 +2213,11 @@ out: #endif -static int get_super_block(struct file_system_type *fs_type, +static struct dentry *get_super_block(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, reiserfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, reiserfs_fill_super); } static int __init init_reiserfs_fs(void) @@ -2253,7 +2252,7 @@ static void __exit exit_reiserfs_fs(void) struct file_system_type reiserfs_fs_type = { .owner = THIS_MODULE, .name = "reiserfs", - .get_sb = get_super_block, + .mount = get_super_block, .kill_sb = reiserfs_kill_sb, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 2685805..6647f90 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -552,20 +552,19 @@ error_rsb: /* * get a superblock for mounting */ -static int romfs_get_sb(struct file_system_type *fs_type, +static struct dentry *romfs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, struct vfsmount *mnt) + void *data) { - int ret = -EINVAL; + struct dentry *ret = ERR_PTR(-EINVAL); #ifdef CONFIG_ROMFS_ON_MTD - ret = get_sb_mtd(fs_type, flags, dev_name, data, romfs_fill_super, - mnt); + ret = mount_mtd(fs_type, flags, dev_name, data, romfs_fill_super); #endif #ifdef CONFIG_ROMFS_ON_BLOCK - if (ret == -EINVAL) - ret = get_sb_bdev(fs_type, flags, dev_name, data, - romfs_fill_super, mnt); + if (ret == ERR_PTR(-EINVAL)) + ret = mount_bdev(fs_type, flags, dev_name, data, + romfs_fill_super); #endif return ret; } @@ -592,7 +591,7 @@ static void romfs_kill_sb(struct super_block *sb) static struct file_system_type romfs_fs_type = { .owner = THIS_MODULE, .name = "romfs", - .get_sb = romfs_get_sb, + .mount = romfs_mount, .kill_sb = romfs_kill_sb, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 07a4f11..24de30b 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -370,12 +370,10 @@ static void squashfs_put_super(struct super_block *sb) } -static int squashfs_get_sb(struct file_system_type *fs_type, int flags, - const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *squashfs_mount(struct file_system_type *fs_type, int flags, + const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, squashfs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, squashfs_fill_super); } @@ -451,7 +449,7 @@ static void squashfs_destroy_inode(struct inode *inode) static struct file_system_type squashfs_fs_type = { .owner = THIS_MODULE, .name = "squashfs", - .get_sb = squashfs_get_sb, + .mount = squashfs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV }; diff --git a/fs/squashfs/xattr.c b/fs/squashfs/xattr.c index 652b854..3876c36 100644 --- a/fs/squashfs/xattr.c +++ b/fs/squashfs/xattr.c @@ -158,17 +158,18 @@ static int squashfs_xattr_get(struct inode *inode, int name_index, strncmp(target, name, name_size) == 0) { /* found xattr */ if (type & SQUASHFS_XATTR_VALUE_OOL) { - __le64 xattr; + __le64 xattr_val; + u64 xattr; /* val is a reference to the real location */ err = squashfs_read_metadata(sb, &val, &start, &offset, sizeof(val)); if (err < 0) goto failed; - err = squashfs_read_metadata(sb, &xattr, &start, - &offset, sizeof(xattr)); + err = squashfs_read_metadata(sb, &xattr_val, + &start, &offset, sizeof(xattr_val)); if (err < 0) goto failed; - xattr = le64_to_cpu(xattr); + xattr = le64_to_cpu(xattr_val); start = SQUASHFS_XATTR_BLK(xattr) + msblk->xattr_table; offset = SQUASHFS_XATTR_OFFSET(xattr); diff --git a/fs/squashfs/xattr.h b/fs/squashfs/xattr.h index 49fe0d7..b634efc 100644 --- a/fs/squashfs/xattr.h +++ b/fs/squashfs/xattr.h @@ -25,7 +25,7 @@ extern __le64 *squashfs_read_xattr_id_table(struct super_block *, u64, u64 *, int *); extern int squashfs_xattr_lookup(struct super_block *, unsigned int, int *, - int *, unsigned long long *); + unsigned int *, unsigned long long *); #else static inline __le64 *squashfs_read_xattr_id_table(struct super_block *sb, u64 start, u64 *xattr_table_start, int *xattr_ids) @@ -35,7 +35,7 @@ static inline __le64 *squashfs_read_xattr_id_table(struct super_block *sb, } static inline int squashfs_xattr_lookup(struct super_block *sb, - unsigned int index, int *count, int *size, + unsigned int index, int *count, unsigned int *size, unsigned long long *xattr) { return 0; diff --git a/fs/squashfs/xattr_id.c b/fs/squashfs/xattr_id.c index cfb4110..d33be5d 100644 --- a/fs/squashfs/xattr_id.c +++ b/fs/squashfs/xattr_id.c @@ -34,6 +34,7 @@ #include "squashfs_fs_sb.h" #include "squashfs_fs_i.h" #include "squashfs.h" +#include "xattr.h" /* * Map xattr id using the xattr id look up table @@ -715,15 +715,14 @@ static int ns_set_super(struct super_block *sb, void *data) return set_anon_super(sb, NULL); } -int get_sb_ns(struct file_system_type *fs_type, int flags, void *data, - int (*fill_super)(struct super_block *, void *, int), - struct vfsmount *mnt) +struct dentry *mount_ns(struct file_system_type *fs_type, int flags, + void *data, int (*fill_super)(struct super_block *, void *, int)) { struct super_block *sb; sb = sget(fs_type, ns_test_super, ns_set_super, data); if (IS_ERR(sb)) - return PTR_ERR(sb); + return ERR_CAST(sb); if (!sb->s_root) { int err; @@ -731,17 +730,16 @@ int get_sb_ns(struct file_system_type *fs_type, int flags, void *data, err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) { deactivate_locked_super(sb); - return err; + return ERR_PTR(err); } sb->s_flags |= MS_ACTIVE; } - simple_set_mnt(mnt, sb); - return 0; + return dget(sb->s_root); } -EXPORT_SYMBOL(get_sb_ns); +EXPORT_SYMBOL(mount_ns); #ifdef CONFIG_BLOCK static int set_bdev_super(struct super_block *s, void *data) @@ -762,10 +760,9 @@ static int test_bdev_super(struct super_block *s, void *data) return (void *)s->s_bdev == data; } -int get_sb_bdev(struct file_system_type *fs_type, +struct dentry *mount_bdev(struct file_system_type *fs_type, int flags, const char *dev_name, void *data, - int (*fill_super)(struct super_block *, void *, int), - struct vfsmount *mnt) + int (*fill_super)(struct super_block *, void *, int)) { struct block_device *bdev; struct super_block *s; @@ -777,7 +774,7 @@ int get_sb_bdev(struct file_system_type *fs_type, bdev = open_bdev_exclusive(dev_name, mode, fs_type); if (IS_ERR(bdev)) - return PTR_ERR(bdev); + return ERR_CAST(bdev); /* * once the super is inserted into the list by sget, s_umount @@ -829,15 +826,30 @@ int get_sb_bdev(struct file_system_type *fs_type, bdev->bd_super = s; } - simple_set_mnt(mnt, s); - return 0; + return dget(s->s_root); error_s: error = PTR_ERR(s); error_bdev: close_bdev_exclusive(bdev, mode); error: - return error; + return ERR_PTR(error); +} +EXPORT_SYMBOL(mount_bdev); + +int get_sb_bdev(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data, + int (*fill_super)(struct super_block *, void *, int), + struct vfsmount *mnt) +{ + struct dentry *root; + + root = mount_bdev(fs_type, flags, dev_name, data, fill_super); + if (IS_ERR(root)) + return PTR_ERR(root); + mnt->mnt_root = root; + mnt->mnt_sb = root->d_sb; + return 0; } EXPORT_SYMBOL(get_sb_bdev); @@ -856,29 +868,42 @@ void kill_block_super(struct super_block *sb) EXPORT_SYMBOL(kill_block_super); #endif -int get_sb_nodev(struct file_system_type *fs_type, +struct dentry *mount_nodev(struct file_system_type *fs_type, int flags, void *data, - int (*fill_super)(struct super_block *, void *, int), - struct vfsmount *mnt) + int (*fill_super)(struct super_block *, void *, int)) { int error; struct super_block *s = sget(fs_type, NULL, set_anon_super, NULL); if (IS_ERR(s)) - return PTR_ERR(s); + return ERR_CAST(s); s->s_flags = flags; error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(s); - return error; + return ERR_PTR(error); } s->s_flags |= MS_ACTIVE; - simple_set_mnt(mnt, s); - return 0; + return dget(s->s_root); } +EXPORT_SYMBOL(mount_nodev); +int get_sb_nodev(struct file_system_type *fs_type, + int flags, void *data, + int (*fill_super)(struct super_block *, void *, int), + struct vfsmount *mnt) +{ + struct dentry *root; + + root = mount_nodev(fs_type, flags, data, fill_super); + if (IS_ERR(root)) + return PTR_ERR(root); + mnt->mnt_root = root; + mnt->mnt_sb = root->d_sb; + return 0; +} EXPORT_SYMBOL(get_sb_nodev); static int compare_single(struct super_block *s, void *p) @@ -886,29 +911,42 @@ static int compare_single(struct super_block *s, void *p) return 1; } -int get_sb_single(struct file_system_type *fs_type, +struct dentry *mount_single(struct file_system_type *fs_type, int flags, void *data, - int (*fill_super)(struct super_block *, void *, int), - struct vfsmount *mnt) + int (*fill_super)(struct super_block *, void *, int)) { struct super_block *s; int error; s = sget(fs_type, compare_single, set_anon_super, NULL); if (IS_ERR(s)) - return PTR_ERR(s); + return ERR_CAST(s); if (!s->s_root) { s->s_flags = flags; error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(s); - return error; + return ERR_PTR(error); } s->s_flags |= MS_ACTIVE; } else { do_remount_sb(s, flags, data, 0); } - simple_set_mnt(mnt, s); + return dget(s->s_root); +} +EXPORT_SYMBOL(mount_single); + +int get_sb_single(struct file_system_type *fs_type, + int flags, void *data, + int (*fill_super)(struct super_block *, void *, int), + struct vfsmount *mnt) +{ + struct dentry *root; + root = mount_single(fs_type, flags, data, fill_super); + if (IS_ERR(root)) + return PTR_ERR(root); + mnt->mnt_root = root; + mnt->mnt_sb = root->d_sb; return 0; } @@ -918,6 +956,7 @@ struct vfsmount * vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void *data) { struct vfsmount *mnt; + struct dentry *root; char *secdata = NULL; int error; @@ -942,9 +981,19 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void goto out_free_secdata; } - error = type->get_sb(type, flags, name, data, mnt); - if (error < 0) - goto out_free_secdata; + if (type->mount) { + root = type->mount(type, flags, name, data); + if (IS_ERR(root)) { + error = PTR_ERR(root); + goto out_free_secdata; + } + mnt->mnt_root = root; + mnt->mnt_sb = root->d_sb; + } else { + error = type->get_sb(type, flags, name, data, mnt); + if (error < 0) + goto out_free_secdata; + } BUG_ON(!mnt->mnt_sb); WARN_ON(!mnt->mnt_sb->s_bdi); mnt->mnt_sb->s_flags |= MS_BORN; diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index f2af225..2668957 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -23,7 +23,7 @@ #include "sysfs.h" -static struct vfsmount *sysfs_mount; +static struct vfsmount *sysfs_mnt; struct kmem_cache *sysfs_dir_cachep; static const struct super_operations sysfs_ops = { @@ -95,18 +95,17 @@ static int sysfs_set_super(struct super_block *sb, void *data) return error; } -static int sysfs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *sysfs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { struct sysfs_super_info *info; enum kobj_ns_type type; struct super_block *sb; int error; - error = -ENOMEM; info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) - goto out; + return ERR_PTR(-ENOMEM); for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) info->ns[type] = kobj_ns_current(type); @@ -114,24 +113,19 @@ static int sysfs_get_sb(struct file_system_type *fs_type, sb = sget(fs_type, sysfs_test_super, sysfs_set_super, info); if (IS_ERR(sb) || sb->s_fs_info != info) kfree(info); - if (IS_ERR(sb)) { - error = PTR_ERR(sb); - goto out; - } + if (IS_ERR(sb)) + return ERR_CAST(sb); if (!sb->s_root) { sb->s_flags = flags; error = sysfs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(sb); - goto out; + return ERR_PTR(error); } sb->s_flags |= MS_ACTIVE; } - simple_set_mnt(mnt, sb); - error = 0; -out: - return error; + return dget(sb->s_root); } static void sysfs_kill_sb(struct super_block *sb) @@ -147,7 +141,7 @@ static void sysfs_kill_sb(struct super_block *sb) static struct file_system_type sysfs_fs_type = { .name = "sysfs", - .get_sb = sysfs_get_sb, + .mount = sysfs_mount, .kill_sb = sysfs_kill_sb, }; @@ -189,11 +183,11 @@ int __init sysfs_init(void) err = register_filesystem(&sysfs_fs_type); if (!err) { - sysfs_mount = kern_mount(&sysfs_fs_type); - if (IS_ERR(sysfs_mount)) { + sysfs_mnt = kern_mount(&sysfs_fs_type); + if (IS_ERR(sysfs_mnt)) { printk(KERN_ERR "sysfs: could not mount!\n"); - err = PTR_ERR(sysfs_mount); - sysfs_mount = NULL; + err = PTR_ERR(sysfs_mnt); + sysfs_mnt = NULL; unregister_filesystem(&sysfs_fs_type); goto out_err; } diff --git a/fs/sysv/super.c b/fs/sysv/super.c index a0b0cda..3d9c62b 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -526,23 +526,22 @@ failed: /* Every kernel module contains stuff like this. */ -static int sysv_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *sysv_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, sysv_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, sysv_fill_super); } -static int v7_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *v7_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, v7_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, v7_fill_super); } static struct file_system_type sysv_fs_type = { .owner = THIS_MODULE, .name = "sysv", - .get_sb = sysv_get_sb, + .mount = sysv_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; @@ -550,7 +549,7 @@ static struct file_system_type sysv_fs_type = { static struct file_system_type v7_fs_type = { .owner = THIS_MODULE, .name = "v7", - .get_sb = v7_get_sb, + .mount = v7_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 9a47c9f..91fac54 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -2038,8 +2038,8 @@ static int sb_test(struct super_block *sb, void *data) return c->vi.cdev == *dev; } -static int ubifs_get_sb(struct file_system_type *fs_type, int flags, - const char *name, void *data, struct vfsmount *mnt) +static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, + const char *name, void *data) { struct ubi_volume_desc *ubi; struct ubi_volume_info vi; @@ -2057,7 +2057,7 @@ static int ubifs_get_sb(struct file_system_type *fs_type, int flags, if (IS_ERR(ubi)) { dbg_err("cannot open \"%s\", error %d", name, (int)PTR_ERR(ubi)); - return PTR_ERR(ubi); + return ERR_CAST(ubi); } ubi_get_volume_info(ubi, &vi); @@ -2095,20 +2095,19 @@ static int ubifs_get_sb(struct file_system_type *fs_type, int flags, /* 'fill_super()' opens ubi again so we must close it here */ ubi_close_volume(ubi); - simple_set_mnt(mnt, sb); - return 0; + return dget(sb->s_root); out_deact: deactivate_locked_super(sb); out_close: ubi_close_volume(ubi); - return err; + return ERR_PTR(err); } static struct file_system_type ubifs_fs_type = { .name = "ubifs", .owner = THIS_MODULE, - .get_sb = ubifs_get_sb, + .mount = ubifs_mount, .kill_sb = kill_anon_super, }; diff --git a/fs/udf/super.c b/fs/udf/super.c index 76f3d6d..4a5c7c6 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -107,17 +107,16 @@ struct logicalVolIntegrityDescImpUse *udf_sb_lvidiu(struct udf_sb_info *sbi) } /* UDF filesystem type */ -static int udf_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, - struct vfsmount *mnt) +static struct dentry *udf_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, udf_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, udf_fill_super); } static struct file_system_type udf_fstype = { .owner = THIS_MODULE, .name = "udf", - .get_sb = udf_get_sb, + .mount = udf_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 6b9be90..2c47dae 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1454,16 +1454,16 @@ static const struct super_operations ufs_super_ops = { .show_options = ufs_show_options, }; -static int ufs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, void *data, struct vfsmount *mnt) +static struct dentry *ufs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, ufs_fill_super, mnt); + return mount_bdev(fs_type, flags, dev_name, data, ufs_fill_super); } static struct file_system_type ufs_fs_type = { .owner = THIS_MODULE, .name = "ufs", - .get_sb = ufs_get_sb, + .mount = ufs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index cf80878..9f3a78f 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1609,16 +1609,14 @@ xfs_fs_fill_super( goto out_free_sb; } -STATIC int -xfs_fs_get_sb( +STATIC struct dentry * +xfs_fs_mount( struct file_system_type *fs_type, int flags, const char *dev_name, - void *data, - struct vfsmount *mnt) + void *data) { - return get_sb_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super, - mnt); + return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super); } static const struct super_operations xfs_super_operations = { @@ -1639,7 +1637,7 @@ static const struct super_operations xfs_super_operations = { static struct file_system_type xfs_fs_type = { .owner = THIS_MODULE, .name = "xfs", - .get_sb = xfs_fs_get_sb, + .mount = xfs_fs_mount, .kill_sb = kill_block_super, .fs_flags = FS_REQUIRES_DEV, }; |
