summaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/afs/dir.c8
-rw-r--r--fs/afs/fsclient.c3
-rw-r--r--fs/afs/inode.c10
-rw-r--r--fs/afs/super.c74
-rw-r--r--fs/afs/write.c21
-rw-r--r--fs/bad_inode.c3
-rw-r--r--fs/btrfs/ctree.c10
-rw-r--r--fs/btrfs/ctree.h15
-rw-r--r--fs/btrfs/delayed-inode.c32
-rw-r--r--fs/btrfs/delayed-inode.h5
-rw-r--r--fs/btrfs/disk-io.c17
-rw-r--r--fs/btrfs/extent-tree.c59
-rw-r--r--fs/btrfs/extent_io.h2
-rw-r--r--fs/btrfs/free-space-cache.c163
-rw-r--r--fs/btrfs/inode.c5
-rw-r--r--fs/btrfs/ioctl.c25
-rw-r--r--fs/btrfs/relocation.c30
-rw-r--r--fs/btrfs/scrub.c69
-rw-r--r--fs/btrfs/super.c2
-rw-r--r--fs/btrfs/sysfs.c146
-rw-r--r--fs/btrfs/transaction.c121
-rw-r--r--fs/btrfs/tree-log.c2
-rw-r--r--fs/btrfs/volumes.c8
-rw-r--r--fs/buffer.c4
-rw-r--r--fs/ceph/addr.c2
-rw-r--r--fs/ceph/caps.c10
-rw-r--r--fs/ceph/dir.c11
-rw-r--r--fs/ceph/export.c4
-rw-r--r--fs/ceph/file.c35
-rw-r--r--fs/ceph/inode.c18
-rw-r--r--fs/ceph/ioctl.c6
-rw-r--r--fs/ceph/locks.c29
-rw-r--r--fs/ceph/snap.c2
-rw-r--r--fs/ceph/xattr.c6
-rw-r--r--fs/cifs/Kconfig3
-rw-r--r--fs/cifs/cache.c6
-rw-r--r--fs/cifs/cifsencrypt.c2
-rw-r--r--fs/cifs/cifsfs.c36
-rw-r--r--fs/cifs/cifsfs.h2
-rw-r--r--fs/cifs/connect.c29
-rw-r--r--fs/cifs/fscache.c51
-rw-r--r--fs/coda/pioctl.c2
-rw-r--r--fs/dcookies.c3
-rw-r--r--fs/exec.c7
-rw-r--r--fs/fat/file.c2
-rw-r--r--fs/fuse/inode.c2
-rw-r--r--fs/gfs2/glock.c9
-rw-r--r--fs/isofs/inode.c3
-rw-r--r--fs/jfs/jfs_logmgr.c2
-rw-r--r--fs/logfs/dir.c8
-rw-r--r--fs/namei.c39
-rw-r--r--fs/nilfs2/btree.c39
-rw-r--r--fs/nilfs2/inode.c7
-rw-r--r--fs/nilfs2/segment.c2
-rw-r--r--fs/ocfs2/super.c2
-rw-r--r--fs/proc/base.c6
-rw-r--r--fs/proc/namespaces.c9
-rw-r--r--fs/proc/proc_sysctl.c3
-rw-r--r--fs/proc/root.c11
-rw-r--r--fs/reiserfs/xattr.c2
-rw-r--r--fs/super.c2
-rw-r--r--fs/sysfs/mount.c37
-rw-r--r--fs/sysfs/sysfs.h2
-rw-r--r--fs/timerfd.c5
-rw-r--r--fs/ubifs/super.c136
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c50
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c75
-rw-r--r--fs/xfs/xfs_log.c11
69 files changed, 832 insertions, 732 deletions
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index 20c106f24927..1b0b19550015 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -584,11 +584,11 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry,
success:
d_add(dentry, inode);
- _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%llu }",
+ _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%u }",
fid.vnode,
fid.unique,
dentry->d_inode->i_ino,
- (unsigned long long)dentry->d_inode->i_version);
+ dentry->d_inode->i_generation);
return NULL;
}
@@ -671,10 +671,10 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
* been deleted and replaced, and the original vnode ID has
* been reused */
if (fid.unique != vnode->fid.unique) {
- _debug("%s: file deleted (uq %u -> %u I:%llu)",
+ _debug("%s: file deleted (uq %u -> %u I:%u)",
dentry->d_name.name, fid.unique,
vnode->fid.unique,
- (unsigned long long)dentry->d_inode->i_version);
+ dentry->d_inode->i_generation);
spin_lock(&vnode->lock);
set_bit(AFS_VNODE_DELETED, &vnode->flags);
spin_unlock(&vnode->lock);
diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c
index 4bd0218473a9..346e3289abd7 100644
--- a/fs/afs/fsclient.c
+++ b/fs/afs/fsclient.c
@@ -89,7 +89,7 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp,
i_size_write(&vnode->vfs_inode, size);
vnode->vfs_inode.i_uid = status->owner;
vnode->vfs_inode.i_gid = status->group;
- vnode->vfs_inode.i_version = vnode->fid.unique;
+ vnode->vfs_inode.i_generation = vnode->fid.unique;
vnode->vfs_inode.i_nlink = status->nlink;
mode = vnode->vfs_inode.i_mode;
@@ -102,6 +102,7 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp,
vnode->vfs_inode.i_ctime.tv_sec = status->mtime_server;
vnode->vfs_inode.i_mtime = vnode->vfs_inode.i_ctime;
vnode->vfs_inode.i_atime = vnode->vfs_inode.i_ctime;
+ vnode->vfs_inode.i_version = data_version;
}
expected_version = status->data_version;
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index db66c5201474..0fdab6e03d87 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -75,7 +75,8 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key)
inode->i_ctime.tv_nsec = 0;
inode->i_atime = inode->i_mtime = inode->i_ctime;
inode->i_blocks = 0;
- inode->i_version = vnode->fid.unique;
+ inode->i_generation = vnode->fid.unique;
+ inode->i_version = vnode->status.data_version;
inode->i_mapping->a_ops = &afs_fs_aops;
/* check to see whether a symbolic link is really a mountpoint */
@@ -100,7 +101,7 @@ static int afs_iget5_test(struct inode *inode, void *opaque)
struct afs_iget_data *data = opaque;
return inode->i_ino == data->fid.vnode &&
- inode->i_version == data->fid.unique;
+ inode->i_generation == data->fid.unique;
}
/*
@@ -122,7 +123,7 @@ static int afs_iget5_set(struct inode *inode, void *opaque)
struct afs_vnode *vnode = AFS_FS_I(inode);
inode->i_ino = data->fid.vnode;
- inode->i_version = data->fid.unique;
+ inode->i_generation = data->fid.unique;
vnode->fid = data->fid;
vnode->volume = data->volume;
@@ -380,8 +381,7 @@ int afs_getattr(struct vfsmount *mnt, struct dentry *dentry,
inode = dentry->d_inode;
- _enter("{ ino=%lu v=%llu }", inode->i_ino,
- (unsigned long long)inode->i_version);
+ _enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation);
generic_fillattr(inode, stat);
return 0;
diff --git a/fs/afs/super.c b/fs/afs/super.c
index fb240e8766d6..356dcf0929e8 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -31,8 +31,8 @@
static void afs_i_init_once(void *foo);
static struct dentry *afs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data);
+static void afs_kill_super(struct super_block *sb);
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);
static int afs_statfs(struct dentry *dentry, struct kstatfs *buf);
@@ -40,7 +40,7 @@ struct file_system_type afs_fs_type = {
.owner = THIS_MODULE,
.name = "afs",
.mount = afs_mount,
- .kill_sb = kill_anon_super,
+ .kill_sb = afs_kill_super,
.fs_flags = 0,
};
@@ -50,7 +50,6 @@ static const struct super_operations afs_super_ops = {
.drop_inode = afs_drop_inode,
.destroy_inode = afs_destroy_inode,
.evict_inode = afs_evict_inode,
- .put_super = afs_put_super,
.show_options = generic_show_options,
};
@@ -282,19 +281,25 @@ static int afs_parse_device_name(struct afs_mount_params *params,
*/
static int afs_test_super(struct super_block *sb, void *data)
{
- struct afs_mount_params *params = data;
+ struct afs_super_info *as1 = data;
struct afs_super_info *as = sb->s_fs_info;
- return as->volume == params->volume;
+ return as->volume == as1->volume;
+}
+
+static int afs_set_super(struct super_block *sb, void *data)
+{
+ sb->s_fs_info = data;
+ return set_anon_super(sb, NULL);
}
/*
* fill in the superblock
*/
-static int afs_fill_super(struct super_block *sb, void *data)
+static int afs_fill_super(struct super_block *sb,
+ struct afs_mount_params *params)
{
- struct afs_mount_params *params = data;
- struct afs_super_info *as = NULL;
+ struct afs_super_info *as = sb->s_fs_info;
struct afs_fid fid;
struct dentry *root = NULL;
struct inode *inode = NULL;
@@ -302,23 +307,13 @@ static int afs_fill_super(struct super_block *sb, void *data)
_enter("");
- /* allocate a superblock info record */
- as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL);
- if (!as) {
- _leave(" = -ENOMEM");
- return -ENOMEM;
- }
-
- afs_get_volume(params->volume);
- as->volume = params->volume;
-
/* fill in the superblock */
sb->s_blocksize = PAGE_CACHE_SIZE;
sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
sb->s_magic = AFS_FS_MAGIC;
sb->s_op = &afs_super_ops;
- sb->s_fs_info = as;
sb->s_bdi = &as->volume->bdi;
+ strlcpy(sb->s_id, as->volume->vlocation->vldb.name, sizeof(sb->s_id));
/* allocate the root inode and dentry */
fid.vid = as->volume->vid;
@@ -326,7 +321,7 @@ static int afs_fill_super(struct super_block *sb, void *data)
fid.unique = 1;
inode = afs_iget(sb, params->key, &fid, NULL, NULL);
if (IS_ERR(inode))
- goto error_inode;
+ return PTR_ERR(inode);
if (params->autocell)
set_bit(AFS_VNODE_AUTOCELL, &AFS_FS_I(inode)->flags);
@@ -342,16 +337,8 @@ static int afs_fill_super(struct super_block *sb, void *data)
_leave(" = 0");
return 0;
-error_inode:
- ret = PTR_ERR(inode);
- inode = NULL;
error:
iput(inode);
- afs_put_volume(as->volume);
- kfree(as);
-
- sb->s_fs_info = NULL;
-
_leave(" = %d", ret);
return ret;
}
@@ -367,6 +354,7 @@ static struct dentry *afs_mount(struct file_system_type *fs_type,
struct afs_volume *vol;
struct key *key;
char *new_opts = kstrdup(options, GFP_KERNEL);
+ struct afs_super_info *as;
int ret;
_enter(",,%s,%p", dev_name, options);
@@ -399,12 +387,22 @@ static struct dentry *afs_mount(struct file_system_type *fs_type,
ret = PTR_ERR(vol);
goto error;
}
- params.volume = vol;
+
+ /* allocate a superblock info record */
+ as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL);
+ if (!as) {
+ ret = -ENOMEM;
+ afs_put_volume(vol);
+ goto error;
+ }
+ as->volume = vol;
/* allocate a deviceless superblock */
- sb = sget(fs_type, afs_test_super, set_anon_super, &params);
+ sb = sget(fs_type, afs_test_super, afs_set_super, as);
if (IS_ERR(sb)) {
ret = PTR_ERR(sb);
+ afs_put_volume(vol);
+ kfree(as);
goto error;
}
@@ -422,16 +420,16 @@ static struct dentry *afs_mount(struct file_system_type *fs_type,
} else {
_debug("reuse");
ASSERTCMP(sb->s_flags, &, MS_ACTIVE);
+ afs_put_volume(vol);
+ kfree(as);
}
- afs_put_volume(params.volume);
afs_put_cell(params.cell);
kfree(new_opts);
_leave(" = 0 [%p]", sb);
return dget(sb->s_root);
error:
- afs_put_volume(params.volume);
afs_put_cell(params.cell);
key_put(params.key);
kfree(new_opts);
@@ -439,18 +437,12 @@ error:
return ERR_PTR(ret);
}
-/*
- * finish the unmounting process on the superblock
- */
-static void afs_put_super(struct super_block *sb)
+static void afs_kill_super(struct super_block *sb)
{
struct afs_super_info *as = sb->s_fs_info;
-
- _enter("");
-
+ kill_anon_super(sb);
afs_put_volume(as->volume);
-
- _leave("");
+ kfree(as);
}
/*
diff --git a/fs/afs/write.c b/fs/afs/write.c
index 789b3afb3423..b806285ff853 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -84,23 +84,21 @@ void afs_put_writeback(struct afs_writeback *wb)
* partly or wholly fill a page that's under preparation for writing
*/
static int afs_fill_page(struct afs_vnode *vnode, struct key *key,
- loff_t pos, unsigned len, struct page *page)
+ loff_t pos, struct page *page)
{
loff_t i_size;
- unsigned eof;
int ret;
+ int len;
- _enter(",,%llu,%u", (unsigned long long)pos, len);
-
- ASSERTCMP(len, <=, PAGE_CACHE_SIZE);
+ _enter(",,%llu", (unsigned long long)pos);
i_size = i_size_read(&vnode->vfs_inode);
- if (pos + len > i_size)
- eof = i_size;
+ if (pos + PAGE_CACHE_SIZE > i_size)
+ len = i_size - pos;
else
- eof = PAGE_CACHE_SIZE;
+ len = PAGE_CACHE_SIZE;
- ret = afs_vnode_fetch_data(vnode, key, 0, eof, page);
+ ret = afs_vnode_fetch_data(vnode, key, pos, len, page);
if (ret < 0) {
if (ret == -ENOENT) {
_debug("got NOENT from server"
@@ -153,9 +151,8 @@ int afs_write_begin(struct file *file, struct address_space *mapping,
*pagep = page;
/* page won't leak in error case: it eventually gets cleaned off LRU */
- if (!PageUptodate(page)) {
- _debug("not up to date");
- ret = afs_fill_page(vnode, key, pos, len, page);
+ if (!PageUptodate(page) && len != PAGE_CACHE_SIZE) {
+ ret = afs_fill_page(vnode, key, index << PAGE_CACHE_SHIFT, page);
if (ret < 0) {
kfree(candidate);
_leave(" = %d [prep]", ret);
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index 9ad2369d9e35..bfcb18feb1df 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -231,9 +231,6 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer,
static int bad_inode_permission(struct inode *inode, int mask, unsigned int flags)
{
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
-
return -EIO;
}
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index d84089349c82..2e667868e0d2 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -1228,6 +1228,7 @@ static void reada_for_search(struct btrfs_root *root,
u32 nr;
u32 blocksize;
u32 nscan = 0;
+ bool map = true;
if (level != 1)
return;
@@ -1249,8 +1250,11 @@ static void reada_for_search(struct btrfs_root *root,
nritems = btrfs_header_nritems(node);
nr = slot;
+ if (node->map_token || path->skip_locking)
+ map = false;
+
while (1) {
- if (!node->map_token) {
+ if (map && !node->map_token) {
unsigned long offset = btrfs_node_key_ptr_offset(nr);
map_private_extent_buffer(node, offset,
sizeof(struct btrfs_key_ptr),
@@ -1277,7 +1281,7 @@ static void reada_for_search(struct btrfs_root *root,
if ((search <= target && target - search <= 65536) ||
(search > target && search - target <= 65536)) {
gen = btrfs_node_ptr_generation(node, nr);
- if (node->map_token) {
+ if (map && node->map_token) {
unmap_extent_buffer(node, node->map_token,
KM_USER1);
node->map_token = NULL;
@@ -1289,7 +1293,7 @@ static void reada_for_search(struct btrfs_root *root,
if ((nread > 65536 || nscan > 32))
break;
}
- if (node->map_token) {
+ if (map && node->map_token) {
unmap_extent_buffer(node, node->map_token, KM_USER1);
node->map_token = NULL;
}
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 378b5b4443f3..300628795fdb 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -967,6 +967,12 @@ struct btrfs_fs_info {
struct srcu_struct subvol_srcu;
spinlock_t trans_lock;
+ /*
+ * the reloc mutex goes with the trans lock, it is taken
+ * during commit to protect us from the relocation code
+ */
+ struct mutex reloc_mutex;
+
struct list_head trans_list;
struct list_head hashers;
struct list_head dead_roots;
@@ -1172,6 +1178,14 @@ struct btrfs_root {
u32 type;
u64 highest_objectid;
+
+ /* btrfs_record_root_in_trans is a multi-step process,
+ * and it can race with the balancing code. But the
+ * race is very small, and only the first time the root
+ * is added to each transaction. So in_trans_setup
+ * is used to tell us when more checks are required
+ */
+ unsigned long in_trans_setup;
int ref_cows;
int track_dirty;
int in_radix;
@@ -1181,7 +1195,6 @@ struct btrfs_root {
struct btrfs_key defrag_max;
int defrag_running;
char *name;
- int in_sysfs;
/* the dirty list is only used by non-reference counted roots */
struct list_head dirty_list;
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 6462c29d2d37..f1cbd028f7b3 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -297,7 +297,6 @@ struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
item->data_len = data_len;
item->ins_or_del = 0;
item->bytes_reserved = 0;
- item->block_rsv = NULL;
item->delayed_node = NULL;
atomic_set(&item->refs, 1);
}
@@ -593,10 +592,8 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
num_bytes = btrfs_calc_trans_metadata_size(root, 1);
ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
- if (!ret) {
+ if (!ret)
item->bytes_reserved = num_bytes;
- item->block_rsv = dst_rsv;
- }
return ret;
}
@@ -604,10 +601,13 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
struct btrfs_delayed_item *item)
{
+ struct btrfs_block_rsv *rsv;
+
if (!item->bytes_reserved)
return;
- btrfs_block_rsv_release(root, item->block_rsv,
+ rsv = &root->fs_info->global_block_rsv;
+ btrfs_block_rsv_release(root, rsv,
item->bytes_reserved);
}
@@ -1014,6 +1014,7 @@ int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
struct btrfs_delayed_root *delayed_root;
struct btrfs_delayed_node *curr_node, *prev_node;
struct btrfs_path *path;
+ struct btrfs_block_rsv *block_rsv;
int ret = 0;
path = btrfs_alloc_path();
@@ -1021,6 +1022,9 @@ int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
return -ENOMEM;
path->leave_spinning = 1;
+ block_rsv = trans->block_rsv;
+ trans->block_rsv = &root->fs_info->global_block_rsv;
+
delayed_root = btrfs_get_delayed_root(root);
curr_node = btrfs_first_delayed_node(delayed_root);
@@ -1045,6 +1049,7 @@ int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
}
btrfs_free_path(path);
+ trans->block_rsv = block_rsv;
return ret;
}
@@ -1052,6 +1057,7 @@ static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
struct btrfs_delayed_node *node)
{
struct btrfs_path *path;
+ struct btrfs_block_rsv *block_rsv;
int ret;
path = btrfs_alloc_path();
@@ -1059,6 +1065,9 @@ static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
return -ENOMEM;
path->leave_spinning = 1;
+ block_rsv = trans->block_rsv;
+ trans->block_rsv = &node->root->fs_info->global_block_rsv;
+
ret = btrfs_insert_delayed_items(trans, path, node->root, node);
if (!ret)
ret = btrfs_delete_delayed_items(trans, path, node->root, node);
@@ -1066,6 +1075,7 @@ static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
ret = btrfs_update_delayed_inode(trans, node->root, path, node);
btrfs_free_path(path);
+ trans->block_rsv = block_rsv;
return ret;
}
@@ -1116,6 +1126,7 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
struct btrfs_path *path;
struct btrfs_delayed_node *delayed_node = NULL;
struct btrfs_root *root;
+ struct btrfs_block_rsv *block_rsv;
unsigned long nr = 0;
int need_requeue = 0;
int ret;
@@ -1134,6 +1145,9 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
if (IS_ERR(trans))
goto free_path;
+ block_rsv = trans->block_rsv;
+ trans->block_rsv = &root->fs_info->global_block_rsv;
+
ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
if (!ret)
ret = btrfs_delete_delayed_items(trans, path, root,
@@ -1176,6 +1190,7 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
nr = trans->blocks_used;
+ trans->block_rsv = block_rsv;
btrfs_end_transaction_dmeta(trans, root);
__btrfs_btree_balance_dirty(root, nr);
free_path:
@@ -1222,6 +1237,13 @@ again:
return 0;
}
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root)
+{
+ struct btrfs_delayed_root *delayed_root;
+ delayed_root = btrfs_get_delayed_root(root);
+ WARN_ON(btrfs_first_delayed_node(delayed_root));
+}
+
void btrfs_balance_delayed_items(struct btrfs_root *root)
{
struct btrfs_delayed_root *delayed_root;
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index eb7d240aa648..d1a6a2915c66 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -75,7 +75,6 @@ struct btrfs_delayed_item {
struct list_head tree_list; /* used for batch insert/delete items */
struct list_head readdir_list; /* used for readdir items */
u64 bytes_reserved;
- struct btrfs_block_rsv *block_rsv;
struct btrfs_delayed_node *delayed_node;
atomic_t refs;
int ins_or_del;
@@ -138,4 +137,8 @@ int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent,
/* for init */
int __init btrfs_delayed_inode_init(void);
void btrfs_delayed_inode_exit(void);
+
+/* for debugging */
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root);
+
#endif
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index a203d363184d..1ac8db5dc0a3 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1044,7 +1044,6 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
root->last_trans = 0;
root->highest_objectid = 0;
root->name = NULL;
- root->in_sysfs = 0;
root->inode_tree = RB_ROOT;
INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
root->block_rsv = NULL;
@@ -1300,19 +1299,21 @@ again:
return root;
root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
- if (!root->free_ino_ctl)
- goto fail;
root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
GFP_NOFS);
- if (!root->free_ino_pinned)
+ if (!root->free_ino_pinned || !root->free_ino_ctl) {
+ ret = -ENOMEM;
goto fail;
+ }
btrfs_init_free_ino_ctl(root);
mutex_init(&root->fs_commit_mutex);
spin_lock_init(&root->cache_lock);
init_waitqueue_head(&root->cache_wait);
- set_anon_super(&root->anon_super, NULL);
+ ret = set_anon_super(&root->anon_super, NULL);
+ if (ret)
+ goto fail;
if (btrfs_root_refs(&root->root_item) == 0) {
ret = -ENOENT;
@@ -1618,6 +1619,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
spin_lock_init(&fs_info->fs_roots_radix_lock);
spin_lock_init(&fs_info->delayed_iput_lock);
spin_lock_init(&fs_info->defrag_inodes_lock);
+ mutex_init(&fs_info->reloc_mutex);
init_completion(&fs_info->kobj_unregister);
fs_info->tree_root = tree_root;
@@ -1668,8 +1670,6 @@ struct btrfs_root *open_ctree(struct super_block *sb,
init_waitqueue_head(&fs_info->scrub_pause_wait);
init_rwsem(&fs_info->scrub_super_lock);
fs_info->scrub_workers_refcnt = 0;
- btrfs_init_workers(&fs_info->scrub_workers, "scrub",
- fs_info->thread_pool_size, &fs_info->generic_worker);
sb->s_blocksize = 4096;
sb->s_blocksize_bits = blksize_bits(4096);
@@ -2911,9 +2911,8 @@ static int btrfs_destroy_delalloc_inodes(struct btrfs_root *root)
INIT_LIST_HEAD(&splice);
- list_splice_init(&root->fs_info->delalloc_inodes, &splice);
-
spin_lock(&root->fs_info->delalloc_lock);
+ list_splice_init(&root->fs_info->delalloc_inodes, &splice);
while (!list_empty(&splice)) {
btrfs_inode = list_entry(splice.next, struct btrfs_inode,
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 5b9b6b6df242..1f61bf5b4960 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3089,6 +3089,13 @@ alloc:
}
goto again;
}
+
+ /*
+ * If we have less pinned bytes than we want to allocate then
+ * don't bother committing the transaction, it won't help us.
+ */
+ if (data_sinfo->bytes_pinned < bytes)
+ committed = 1;
spin_unlock(&data_sinfo->lock);
/* commit the current transaction and try again */
@@ -3307,10 +3314,6 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans,
if (reserved == 0)
return 0;
- /* nothing to shrink - nothing to reclaim */
- if (root->fs_info->delalloc_bytes == 0)
- return 0;
-
max_reclaim = min(reserved, to_reclaim);
while (loops < 1024) {
@@ -5211,9 +5214,7 @@ loop:
* LOOP_NO_EMPTY_SIZE, set empty_size and empty_cluster to 0 and try
* again
*/
- if (!ins->objectid && loop < LOOP_NO_EMPTY_SIZE &&
- (found_uncached_bg || empty_size || empty_cluster ||
- allowed_chunk_alloc)) {
+ if (!ins->objectid && loop < LOOP_NO_EMPTY_SIZE) {
index = 0;
if (loop == LOOP_FIND_IDEAL && found_uncached_bg) {
found_uncached_bg = false;
@@ -5253,32 +5254,36 @@ loop:
goto search;
}
- if (loop < LOOP_CACHING_WAIT) {
- loop++;
- goto search;
- }
+ loop++;
if (loop == LOOP_ALLOC_CHUNK) {
- empty_size = 0;
- empty_cluster = 0;
- }
+ if (allowed_chunk_alloc) {
+ ret = do_chunk_alloc(trans, root, num_bytes +
+ 2 * 1024 * 1024, data,
+ CHUNK_ALLOC_LIMITED);
+ allowed_chunk_alloc = 0;
+ if (ret == 1)
+ done_chunk_alloc = 1;
+ } else if (!done_chunk_alloc &&
+ space_info->force_alloc ==
+ CHUNK_ALLOC_NO_FORCE) {
+ space_info->force_alloc = CHUNK_ALLOC_LIMITED;
+ }
- if (allowed_chunk_alloc) {
- ret = do_chunk_alloc(trans, root, num_bytes +
- 2 * 1024 * 1024, data,
- CHUNK_ALLOC_LIMITED);
- allowed_chunk_alloc = 0;
- done_chunk_alloc = 1;
- } else if (!done_chunk_alloc &&
- space_info->force_alloc == CHUNK_ALLOC_NO_FORCE) {
- space_info->force_alloc = CHUNK_ALLOC_LIMITED;
+ /*
+ * We didn't allocate a chunk, go ahead and drop the
+ * empty size and loop again.
+ */
+ if (!done_chunk_alloc)
+ loop = LOOP_NO_EMPTY_SIZE;
}
- if (loop < LOOP_NO_EMPTY_SIZE) {
- loop++;
- goto search;
+ if (loop == LOOP_NO_EMPTY_SIZE) {
+ empty_size = 0;
+ empty_cluster = 0;
}
- ret = -ENOSPC;
+
+ goto search;
} else if (!ins->objectid) {
ret = -ENOSPC;
} else if (ins->objectid) {
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 4e8445a4757c..a11a92ee2d30 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -126,9 +126,9 @@ struct extent_buffer {
unsigned long map_len;
struct page *first_page;
unsigned long bflags;
- atomic_t refs;
struct list_head leak_list;
struct rcu_head rcu_head;
+ atomic_t refs;
/* the spinlock is used to protect most operations */
spinlock_t lock;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index ad144736a5fd..9f985a429877 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -250,7 +250,7 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
pgoff_t index = 0;
unsigned long first_page_offset;
int num_checksums;
- int ret = 0, ret2;
+ int ret = 0;
INIT_LIST_HEAD(&bitmaps);
@@ -421,11 +421,10 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
goto free_cache;
}
spin_lock(&ctl->tree_lock);
- ret2 = link_free_space(ctl, e);
+ ret = link_free_space(ctl, e);
ctl->total_bitmaps++;
ctl->op->recalc_thresholds(ctl);
spin_unlock(&ctl->tree_lock);
- list_add_tail(&e->list, &bitmaps);
if (ret) {
printk(KERN_ERR "Duplicate entries in "
"free space cache, dumping\n");
@@ -434,6 +433,7 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode,
page_cache_release(page);
goto free_cache;
}
+ list_add_tail(&e->list, &bitmaps);
}
num_entries--;
@@ -1417,6 +1417,23 @@ again:
return 0;
}
+static u64 add_bytes_to_bitmap(struct btrfs_free_space_ctl *ctl,
+ struct btrfs_free_space *info, u64 offset,
+ u64 bytes)
+{
+ u64 bytes_to_set = 0;
+ u64 end;
+
+ end = info->offset + (u64)(BITS_PER_BITMAP * ctl->unit);
+
+ bytes_to_set = min(end - offset, bytes);
+
+ bitmap_set_bits(ctl, info, offset, bytes_to_set);
+
+ return bytes_to_set;
+
+}
+
static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
struct btrfs_free_space *info)
{
@@ -1453,12 +1470,18 @@ static bool use_bitmap(struct btrfs_free_space_ctl *ctl,
return true;
}
+static struct btrfs_free_space_op free_space_op = {
+ .recalc_thresholds = recalculate_thresholds,
+ .use_bitmap = use_bitmap,
+};
+
static int insert_into_bitmap(struct btrfs_free_space_ctl *ctl,
struct btrfs_free_space *info)
{
struct btrfs_free_space *bitmap_info;
+ struct btrfs_block_group_cache *block_group = NULL;
int added = 0;
- u64 bytes, offset, end;
+ u64 bytes, offset, bytes_added;
int ret;
bytes = info->bytes;
@@ -1467,7 +1490,49 @@ static int insert_into_bitmap(struct btrfs_free_space_ctl *ctl,
if (!ctl->op->use_bitmap(ctl, info))
return 0;
+ if (ctl->op == &free_space_op)
+ block_group = ctl->private;
again:
+ /*
+ * Since we link bitmaps right into the cluster we need to see if we
+ * have a cluster here, and if so and it has our bitmap we need to add
+ * the free space to that bitmap.
+ */
+ if (block_group && !list_empty(&block_group->cluster_list)) {
+ struct btrfs_free_cluster *cluster;
+ struct rb_node *node;
+ struct btrfs_free_space *entry;
+
+ cluster = list_entry(block_group->cluster_list.next,
+ struct btrfs_free_cluster,
+ block_group_list);
+ spin_lock(&cluster->lock);
+ node = rb_first(&cluster->root);
+ if (!node) {
+ spin_unlock(&cluster->lock);
+ goto no_cluster_bitmap;
+ }
+
+ entry = rb_entry(node, struct btrfs_free_space, offset_index);
+ if (!entry->bitmap) {
+ spin_unlock(&cluster->lock);
+ goto no_cluster_bitmap;
+ }
+
+ if (entry->offset == offset_to_bitmap(ctl, offset)) {
+ bytes_added = add_bytes_to_bitmap(ctl, entry,
+ offset, bytes);
+ bytes -= bytes_added;
+ offset += bytes_added;
+ }
+ spin_unlock(&cluster->lock);
+ if (!bytes) {
+ ret = 1;
+ goto out;
+ }
+ }
+
+no_cluster_bitmap:
bitmap_info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset),
1, 0);
if (!bitmap_info) {
@@ -1475,19 +1540,10 @@ again:
goto new_bitmap;
}
- end = bitmap_info->offset + (u64)(BITS_PER_BITMAP * ctl->unit);
-
- if (offset >= bitmap_info->offset && offset + bytes > end) {
- bitmap_set_bits(ctl, bitmap_info, offset, end - offset);
- bytes -= end - offset;
- offset = end;
- added = 0;
- } else if (offset >= bitmap_info->offset && offset + bytes <= end) {
- bitmap_set_bits(ctl, bitmap_info, offset, bytes);
- bytes = 0;
- } else {
- BUG();
- }
+ bytes_added = add_bytes_to_bitmap(ctl, bitmap_info, offset, bytes);
+ bytes -= bytes_added;
+ offset += bytes_added;
+ added = 0;
if (!bytes) {
ret = 1;
@@ -1766,11 +1822,6 @@ void btrfs_dump_free_space(struct btrfs_block_group_cache *block_group,
"\n", count);
}
-static struct btrfs_free_space_op free_space_op = {
- .recalc_thresholds = recalculate_thresholds,
- .use_bitmap = use_bitmap,
-};
-
void btrfs_init_free_space_ctl(struct btrfs_block_group_cache *block_group)
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
@@ -2142,9 +2193,11 @@ again:
/*
* This searches the block group for just extents to fill the cluster with.
*/
-static int setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
- struct btrfs_free_cluster *cluster,
- u64 offset, u64 bytes, u64 min_bytes)
+static noinline int
+setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
+ struct btrfs_free_cluster *cluster,
+ struct list_head *bitmaps, u64 offset, u64 bytes,
+ u64 min_bytes)
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
struct btrfs_free_space *first = NULL;
@@ -2166,6 +2219,8 @@ static int setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
* extent entry.
*/
while (entry->bitmap) {
+ if (list_empty(&entry->list))
+ list_add_tail(&entry->list, bitmaps);
node = rb_next(&entry->offset_index);
if (!node)
return -ENOSPC;
@@ -2185,8 +2240,12 @@ static int setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
return -ENOSPC;
entry = rb_entry(node, struct btrfs_free_space, offset_index);
- if (entry->bitmap)
+ if (entry->bitmap) {
+ if (list_empty(&entry->list))
+ list_add_tail(&entry->list, bitmaps);
continue;
+ }
+
/*
* we haven't filled the empty size and the window is
* very large. reset and try again
@@ -2238,9 +2297,11 @@ static int setup_cluster_no_bitmap(struct btrfs_block_group_cache *block_group,
* This specifically looks for bitmaps that may work in the cluster, we assume
* that we have already failed to find extents that will work.
*/
-static int setup_cluster_bitmap(struct btrfs_block_group_cache *block_group,
- struct btrfs_free_cluster *cluster,
- u64 offset, u64 bytes, u64 min_bytes)
+static noinline int
+setup_cluster_bitmap(struct btrfs_block_group_cache *block_group,
+ struct btrfs_free_cluster *cluster,
+ struct list_head *bitmaps, u64 offset, u64 bytes,
+ u64 min_bytes)
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
struct btrfs_free_space *entry;
@@ -2250,10 +2311,39 @@ static int setup_cluster_bitmap(struct btrfs_block_group_cache *block_group,
if (ctl->total_bitmaps == 0)
return -ENOSPC;
+ /*
+ * First check our cached list of bitmaps and see if there is an entry
+ * here that will work.
+ */
+ list_for_each_entry(entry, bitmaps, list) {
+ if (entry->bytes < min_bytes)
+ continue;
+ ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset,
+ bytes, min_bytes);
+ if (!ret)
+ return 0;
+ }
+
+ /*
+ * If we do have entries on our list and we are here then we didn't find
+ * anything, so go ahead and get the next entry after the last entry in
+ * this list and start the search from there.
+ */
+ if (!list_empty(bitmaps)) {
+ entry = list_entry(bitmaps->prev, struct btrfs_free_space,
+ list);
+ node = rb_next(&entry->offset_index);
+ if (!node)
+ return -ENOSPC;
+ entry = rb_entry(node, struct btrfs_free_space, offset_index);
+ goto search;
+ }
+
entry = tree_search_offset(ctl, offset_to_bitmap(ctl, offset), 0, 1);
if (!entry)
return -ENOSPC;
+search:
node = &entry->offset_index;
do {
entry = rb_entry(node, struct btrfs_free_space, offset_index);
@@ -2284,6 +2374,8 @@ int btrfs_find_space_cluster(struct btrfs_trans_handle *trans,
u64 offset, u64 bytes, u64 empty_size)
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
+ struct list_head bitmaps;
+ struct btrfs_free_space *entry, *tmp;
u64 min_bytes;
int ret;
@@ -2322,11 +2414,16 @@ int btrfs_find_space_cluster(struct btrfs_trans_handle *trans,
goto out;
}
- ret = setup_cluster_no_bitmap(block_group, cluster, offset, bytes,
- min_bytes);
+ INIT_LIST_HEAD(&bitmaps);
+ ret = setup_cluster_no_bitmap(block_group, cluster, &bitmaps, offset,
+ bytes, min_bytes);
if (ret)
- ret = setup_cluster_bitmap(block_group, cluster, offset,
- bytes, min_bytes);
+ ret = setup_cluster_bitmap(block_group, cluster, &bitmaps,
+ offset, bytes, min_bytes);
+
+ /* Clear our temporary list */
+ list_for_each_entry_safe(entry, tmp, &bitmaps, list)
+ list_del_init(&entry->list);
if (!ret) {
atomic_inc(&block_group->count);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index ebf95f7a44d6..0a9b10c5b0a7 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1986,7 +1986,7 @@ static int btrfs_readpage_end_io_hook(struct page *page, u64 start, u64 end,
}
if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)
- return 0;
+ goto good;
if (root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID &&
test_range_bit(io_tree, start, end, EXTENT_NODATASUM, 1, NULL)) {
@@ -3076,6 +3076,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
ret = btrfs_update_inode(trans, root, dir);
BUG_ON(ret);
+ btrfs_free_path(path);
return 0;
}
@@ -3646,7 +3647,7 @@ void btrfs_evict_inode(struct inode *inode)
btrfs_i_size_write(inode, 0);
while (1) {
- trans = btrfs_start_transaction(root, 0);
+ trans = btrfs_join_transaction(root);
BUG_ON(IS_ERR(trans));
trans->block_rsv = root->orphan_block_rsv;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index ac37040e426a..a3c4751e07db 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -482,8 +482,10 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
BUG_ON(ret);
+ spin_lock(&root->fs_info->trans_lock);
list_add(&pending_snapshot->list,
&trans->transaction->pending_snapshots);
+ spin_unlock(&root->fs_info->trans_lock);
if (async_transid) {
*async_transid = trans->transid;
ret = btrfs_commit_transaction_async(trans,
@@ -2054,29 +2056,34 @@ static long btrfs_ioctl_rm_dev(struct btrfs_root *root, void __user *arg)
static long btrfs_ioctl_fs_info(struct btrfs_root *root, void __user *arg)
{
- struct btrfs_ioctl_fs_info_args fi_args;
+ struct btrfs_ioctl_fs_info_args *fi_args;
struct btrfs_device *device;
struct btrfs_device *next;
struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
+ int ret = 0;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- fi_args.num_devices = fs_devices->num_devices;
- fi_args.max_id = 0;
- memcpy(&fi_args.fsid, root->fs_info->fsid, sizeof(fi_args.fsid));
+ fi_args = kzalloc(sizeof(*fi_args), GFP_KERNEL);
+ if (!fi_args)
+ return -ENOMEM;
+
+ fi_args->num_devices = fs_devices->num_devices;
+ memcpy(&fi_args->fsid, root->fs_info->fsid, sizeof(fi_args->fsid));
mutex_lock(&fs_devices->device_list_mutex);
list_for_each_entry_safe(device, next, &fs_devices->devices, dev_list) {
- if (device->devid > fi_args.max_id)
- fi_args.max_id = device->devid;
+ if (device->devid > fi_args->max_id)
+ fi_args->max_id = device->devid;
}
mutex_unlock(&fs_devices->device_list_mutex);
- if (copy_to_user(arg, &fi_args, sizeof(fi_args)))
- return -EFAULT;
+ if (copy_to_user(arg, fi_args, sizeof(*fi_args)))
+ ret = -EFAULT;
- return 0;
+ kfree(fi_args);
+ return ret;
}
static long btrfs_ioctl_dev_info(struct btrfs_root *root, void __user *arg)
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index b1ef27cc673b..5e0a3dc79a45 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1368,7 +1368,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
int ret;
if (!root->reloc_root)
- return 0;
+ goto out;
reloc_root = root->reloc_root;
root_item = &reloc_root->root_item;
@@ -1390,6 +1390,8 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
ret = btrfs_update_root(trans, root->fs_info->tree_root,
&reloc_root->root_key, root_item);
BUG_ON(ret);
+
+out:
return 0;
}
@@ -2142,10 +2144,11 @@ int prepare_to_merge(struct reloc_control *rc, int err)
u64 num_bytes = 0;
int ret;
- spin_lock(&root->fs_info->trans_lock);
+ mutex_lock(&root->fs_info->reloc_mutex);
rc->merging_rsv_size += root->nodesize * (BTRFS_MAX_LEVEL - 1) * 2;
rc->merging_rsv_size += rc->nodes_relocated * 2;
- spin_unlock(&root->fs_info->trans_lock);
+ mutex_unlock(&root->fs_info->reloc_mutex);
+
again:
if (!err) {
num_bytes = rc->merging_rsv_size;
@@ -2214,9 +2217,16 @@ int merge_reloc_roots(struct reloc_control *rc)
int ret;
again:
root = rc->extent_root;
- spin_lock(&root->fs_info->trans_lock);
+
+ /*
+ * this serializes us with btrfs_record_root_in_transaction,
+ * we have to make sure nobody is in the middle of
+ * adding their roots to the list while we are
+ * doing this splice
+ */
+ mutex_lock(&root->fs_info->reloc_mutex);
list_splice_init(&rc->reloc_roots, &reloc_roots);
- spin_unlock(&root->fs_info->trans_lock);
+ mutex_unlock(&root->fs_info->reloc_mutex);
while (!list_empty(&reloc_roots)) {
found = 1;
@@ -3590,17 +3600,19 @@ next:
static void set_reloc_control(struct reloc_control *rc)
{
struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
- spin_lock(&fs_info->trans_lock);
+
+ mutex_lock(&fs_info->reloc_mutex);
fs_info->reloc_ctl = rc;
- spin_unlock(&fs_info->trans_lock);
+ mutex_unlock(&fs_info->reloc_mutex);
}
static void unset_reloc_control(struct reloc_control *rc)
{
struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
- spin_lock(&fs_info->trans_lock);
+
+ mutex_lock(&fs_info->reloc_mutex);
fs_info->reloc_ctl = NULL;
- spin_unlock(&fs_info->trans_lock);
+ mutex_unlock(&fs_info->reloc_mutex);
}
static int check_extent_flags(u64 flags)
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index df50fd1eca8f..a8d03d5efb5d 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs/scrub.c
@@ -16,13 +16,7 @@
* Boston, MA 021110-1307, USA.
*/
-#include <linux/sched.h>
-#include <linux/pagemap.h>
-#include <linux/writeback.h>
#include <linux/blkdev.h>
-#include <linux/rbtree.h>
-#include <linux/slab.h>
-#include <linux/workqueue.h>
#include "ctree.h"
#include "volumes.h"
#include "disk-io.h"
@@ -804,18 +798,12 @@ static noinline_for_stack int scrub_stripe(struct scrub_dev *sdev,
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
- goto out;
-
- l = path->nodes[0];
- slot = path->slots[0];
- btrfs_item_key_to_cpu(l, &key, slot);
- if (key.objectid != logical) {
- ret = btrfs_previous_item(root, path, 0,
- BTRFS_EXTENT_ITEM_KEY);
- if (ret < 0)
- goto out;
- }
+ goto out_noplug;
+ /*
+ * we might miss half an extent here, but that doesn't matter,
+ * as it's only the prefetch
+ */
while (1) {
l = path->nodes[0];
slot = path->slots[0];
@@ -824,7 +812,7 @@ static noinline_for_stack int scrub_stripe(struct scrub_dev *sdev,
if (ret == 0)
continue;
if (ret < 0)
- goto out;
+ goto out_noplug;
break;
}
@@ -906,15 +894,20 @@ again:
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
goto out;
-
- l = path->nodes[0];
- slot = path->slots[0];
- btrfs_item_key_to_cpu(l, &key, slot);
- if (key.objectid != logical) {
+ if (ret > 0) {
ret = btrfs_previous_item(root, path, 0,
BTRFS_EXTENT_ITEM_KEY);
if (ret < 0)
goto out;
+ if (ret > 0) {
+ /* there's no smaller item, so stick with the
+ * larger one */
+ btrfs_release_path(path);
+ ret = btrfs_search_slot(NULL, root, &key,
+ path, 0, 0);
+ if (ret < 0)
+ goto out;
+ }
}
while (1) {
@@ -989,6 +982,7 @@ next:
out:
blk_finish_plug(&plug);
+out_noplug:
btrfs_free_path(path);
return ret < 0 ? ret : 0;
}
@@ -1064,8 +1058,15 @@ int scrub_enumerate_chunks(struct scrub_dev *sdev, u64 start, u64 end)
while (1) {
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
- goto out;
- ret = 0;
+ break;
+ if (ret > 0) {
+ if (path->slots[0] >=
+ btrfs_header_nritems(path->nodes[0])) {
+ ret = btrfs_next_leaf(root, path);
+ if (ret)
+ break;
+ }
+ }
l = path->nodes[0];
slot = path->slots[0];
@@ -1075,7 +1076,7 @@ int scrub_enumerate_chunks(struct scrub_dev *sdev, u64 start, u64 end)
if (found_key.objectid != sdev->dev->devid)
break;
- if (btrfs_key_type(&key) != BTRFS_DEV_EXTENT_KEY)
+ if (btrfs_key_type(&found_key) != BTRFS_DEV_EXTENT_KEY)
break;
if (found_key.offset >= end)
@@ -1104,7 +1105,7 @@ int scrub_enumerate_chunks(struct scrub_dev *sdev, u64 start, u64 end)
cache = btrfs_lookup_block_group(fs_info, chunk_offset);
if (!cache) {
ret = -ENOENT;
- goto out;
+ break;
}
ret = scrub_chunk(sdev, chunk_tree, chunk_objectid,
chunk_offset, length);
@@ -1116,9 +1117,13 @@ int scrub_enumerate_chunks(struct scrub_dev *sdev, u64 start, u64 end)
btrfs_release_path(path);
}
-out:
btrfs_free_path(path);
- return ret;
+
+ /*
+ * ret can still be 1 from search_slot or next_leaf,
+ * that's not an error
+ */
+ return ret < 0 ? ret : 0;
}
static noinline_for_stack int scrub_supers(struct scrub_dev *sdev)
@@ -1155,8 +1160,12 @@ static noinline_for_stack int scrub_workers_get(struct btrfs_root *root)
struct btrfs_fs_info *fs_info = root->fs_info;
mutex_lock(&fs_info->scrub_lock);
- if (fs_info->scrub_workers_refcnt == 0)
+ if (fs_info->scrub_workers_refcnt == 0) {
+ btrfs_init_workers(&fs_info->scrub_workers, "scrub",
+ fs_info->thread_pool_size, &fs_info->generic_worker);
+ fs_info->scrub_workers.idle_thresh = 4;
btrfs_start_workers(&fs_info->scrub_workers, 1);
+ }
++fs_info->scrub_workers_refcnt;
mutex_unlock(&fs_info->scrub_lock);
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 117e74e3604b..0bb4ebbb71b7 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -825,7 +825,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
} else {
char b[BDEVNAME_SIZE];
- s->s_flags = flags;
+ s->s_flags = flags | MS_NOSEC;
strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id));
error = btrfs_fill_super(s, fs_devices, data,
flags & MS_SILENT ? 1 : 0);
diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c
index c3c223ae6691..daac9ae6d731 100644
--- a/fs/btrfs/sysfs.c
+++ b/fs/btrfs/sysfs.c
@@ -28,152 +28,6 @@
#include "disk-io.h"
#include "transaction.h"
-static ssize_t root_blocks_used_show(struct btrfs_root *root, char *buf)
-{
- return snprintf(buf, PAGE_SIZE, "%llu\n",
- (unsigned long long)btrfs_root_used(&root->root_item));
-}
-
-static ssize_t root_block_limit_show(struct btrfs_root *root, char *buf)
-{
- return snprintf(buf, PAGE_SIZE, "%llu\n",
- (unsigned long long)btrfs_root_limit(&root->root_item));
-}
-
-static ssize_t super_blocks_used_show(struct btrfs_fs_info *fs, char *buf)
-{
-
- return snprintf(buf, PAGE_SIZE, "%llu\n",
- (unsigned long long)btrfs_super_bytes_used(&fs->super_copy));
-}
-
-static ssize_t super_total_blocks_show(struct btrfs_fs_info *fs, char *buf)
-{
- return snprintf(buf, PAGE_SIZE, "%llu\n",
- (unsigned long long)btrfs_super_total_bytes(&fs->super_copy));
-}
-
-static ssize_t super_blocksize_show(struct btrfs_fs_info *fs, char *buf)
-{
- return snprintf(buf, PAGE_SIZE, "%llu\n",
- (unsigned long long)btrfs_super_sectorsize(&fs->super_copy));
-}
-
-/* this is for root attrs (subvols/snapshots) */
-struct btrfs_root_attr {
- struct attribute attr;
- ssize_t (*show)(struct btrfs_root *, char *);
- ssize_t (*store)(struct btrfs_root *, const char *, size_t);
-};
-
-#define ROOT_ATTR(name, mode, show, store) \
-static struct btrfs_root_attr btrfs_root_attr_##name = __ATTR(name, mode, \
- show, store)
-
-ROOT_ATTR(blocks_used, 0444, root_blocks_used_show, NULL);
-ROOT_ATTR(block_limit, 0644, root_block_limit_show, NULL);
-
-static struct attribute *btrfs_root_attrs[] = {
- &btrfs_root_attr_blocks_used.attr,
- &btrfs_root_attr_block_limit.attr,
- NULL,
-};
-
-/* this is for super attrs (actual full fs) */
-struct btrfs_super_attr {
- struct attribute attr;
- ssize_t (*show)(struct btrfs_fs_info *, char *);
- ssize_t (*store)(struct btrfs_fs_info *, const char *, size_t);
-};
-
-#define SUPER_ATTR(name, mode, show, store) \
-static struct btrfs_super_attr btrfs_super_attr_##name = __ATTR(name, mode, \
- show, store)
-
-SUPER_ATTR(blocks_used, 0444, super_blocks_used_show, NULL);
-SUPER_ATTR(total_blocks, 0444, super_total_blocks_show, NULL);
-SUPER_ATTR(blocksize, 0444, super_blocksize_show, NULL);
-
-static struct attribute *btrfs_super_attrs[] = {
- &btrfs_super_attr_blocks_used.attr,
- &btrfs_super_attr_total_blocks.attr,
- &btrfs_super_attr_blocksize.attr,
- NULL,
-};
-
-static ssize_t btrfs_super_attr_show(struct kobject *kobj,
- struct attribute *attr, char *buf)
-{
- struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
- super_kobj);
- struct btrfs_super_attr *a = container_of(attr,
- struct btrfs_super_attr,
- attr);
-
- return a->show ? a->show(fs, buf) : 0;
-}
-
-static ssize_t btrfs_super_attr_store(struct kobject *kobj,
- struct attribute *attr,
- const char *buf, size_t len)
-{
- struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
- super_kobj);
- struct btrfs_super_attr *a = container_of(attr,
- struct btrfs_super_attr,
- attr);
-
- return a->store ? a->store(fs, buf, len) : 0;
-}
-
-static ssize_t btrfs_root_attr_show(struct kobject *kobj,
- struct attribute *attr, char *buf)
-{
- struct btrfs_root *root = container_of(kobj, struct btrfs_root,
- root_kobj);
- struct btrfs_root_attr *a = container_of(attr,
- struct btrfs_root_attr,
- attr);
-
- return a->show ? a->show(root, buf) : 0;
-}
-
-static ssize_t btrfs_root_attr_store(struct kobject *kobj,
- struct attribute *attr,
- const char *buf, size_t len)
-{
- struct btrfs_root *root = container_of(kobj, struct btrfs_root,
- root_kobj);
- struct btrfs_root_attr *a = container_of(attr,
- struct btrfs_root_attr,
- attr);
- return a->store ? a->store(root, buf, len) : 0;
-}
-
-static void btrfs_super_release(struct kobject *kobj)
-{
- struct btrfs_fs_info *fs = container_of(kobj, struct btrfs_fs_info,
- super_kobj);
- complete(&fs->kobj_unregister);
-}
-
-static void btrfs_root_release(struct kobject *kobj)
-{
- struct btrfs_root *root = container_of(kobj, struct btrfs_root,
- root_kobj);
- complete(&root->kobj_unregister);
-}
-
-static const struct sysfs_ops btrfs_super_attr_ops = {
- .show = btrfs_super_attr_show,
- .store = btrfs_super_attr_store,
-};
-
-static const struct sysfs_ops btrfs_root_attr_ops = {
- .show = btrfs_root_attr_show,
- .store = btrfs_root_attr_store,
-};
-
/* /sys/fs/btrfs/ entry */
static struct kset *btrfs_kset;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index dd719662340e..51dcec86757f 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -126,28 +126,85 @@ static noinline int join_transaction(struct btrfs_root *root, int nofail)
* to make sure the old root from before we joined the transaction is deleted
* when the transaction commits
*/
-int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
+static int record_root_in_trans(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
if (root->ref_cows && root->last_trans < trans->transid) {
WARN_ON(root == root->fs_info->extent_root);
WARN_ON(root->commit_root != root->node);
+ /*
+ * see below for in_trans_setup usage rules
+ * we have the reloc mutex held now, so there
+ * is only one writer in this function
+ */
+ root->in_trans_setup = 1;
+
+ /* make sure readers find in_trans_setup before
+ * they find our root->last_trans update
+ */
+ smp_wmb();
+
spin_lock(&root->fs_info->fs_roots_radix_lock);
if (root->last_trans == trans->transid) {
spin_unlock(&root->fs_info->fs_roots_radix_lock);
return 0;
}
- root->last_trans = trans->transid;
radix_tree_tag_set(&root->fs_info->fs_roots_radix,
(unsigned long)root->root_key.objectid,
BTRFS_ROOT_TRANS_TAG);
spin_unlock(&root->fs_info->fs_roots_radix_lock);
+ root->last_trans = trans->transid;
+
+ /* this is pretty tricky. We don't want to
+ * take the relocation lock in btrfs_record_root_in_trans
+ * unless we're really doing the first setup for this root in
+ * this transaction.
+ *
+ * Normally we'd use root->last_trans as a flag to decide
+ * if we want to take the expensive mutex.
+ *
+ * But, we have to set root->last_trans before we
+ * init the relocation root, otherwise, we trip over warnings
+ * in ctree.c. The solution used here is to flag ourselves
+ * with root->in_trans_setup. When this is 1, we're still
+ * fixing up the reloc trees and everyone must wait.
+ *
+ * When this is zero, they can trust root->last_trans and fly
+ * through btrfs_record_root_in_trans without having to take the
+ * lock. smp_wmb() makes sure that all the writes above are
+ * done before we pop in the zero below
+ */
btrfs_init_reloc_root(trans, root);
+ smp_wmb();
+ root->in_trans_setup = 0;
}
return 0;
}
+
+int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root)
+{
+ if (!root->ref_cows)
+ return 0;
+
+ /*
+ * see record_root_in_trans for comments about in_trans_setup usage
+ * and barriers
+ */
+ smp_rmb();
+ if (root->last_trans == trans->transid &&
+ !root->in_trans_setup)
+ return 0;
+
+ mutex_lock(&root->fs_info->reloc_mutex);
+ record_root_in_trans(trans, root);
+ mutex_unlock(&root->fs_info->reloc_mutex);
+
+ return 0;
+}
+
/* wait for commit against the current transaction to become unblocked
* when this is done, it is safe to start a new transaction, but the current
* transaction might not be fully on disk.
@@ -349,7 +406,7 @@ int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid)
list) {
if (t->in_commit) {
if (t->commit_done)
- goto out;
+ break;
cur_trans = t;
atomic_inc(&cur_trans->use_count);
break;
@@ -882,7 +939,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
parent = dget_parent(dentry);
parent_inode = parent->d_inode;
parent_root = BTRFS_I(parent_inode)->root;
- btrfs_record_root_in_trans(trans, parent_root);
+ record_root_in_trans(trans, parent_root);
/*
* insert the directory item
@@ -900,7 +957,16 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
ret = btrfs_update_inode(trans, parent_root, parent_inode);
BUG_ON(ret);
- btrfs_record_root_in_trans(trans, root);
+ /*
+ * pull in the delayed directory update
+ * and the delayed inode item
+ * otherwise we corrupt the FS during
+ * snapshot
+ */
+ ret = btrfs_run_delayed_items(trans, root);
+ BUG_ON(ret);
+
+ record_root_in_trans(trans, root);
btrfs_set_root_last_snapshot(&root->root_item, trans->transid);
memcpy(new_root_item, &root->root_item, sizeof(*new_root_item));
btrfs_check_and_init_root_item(new_root_item);
@@ -961,14 +1027,6 @@ static noinline int create_pending_snapshots(struct btrfs_trans_handle *trans,
int ret;
list_for_each_entry(pending, head, list) {
- /*
- * We must deal with the delayed items before creating
- * snapshots, or we will create a snapthot with inconsistent
- * information.
- */
- ret = btrfs_run_delayed_items(trans, fs_info->fs_root);
- BUG_ON(ret);
-
ret = create_pending_snapshot(trans, fs_info, pending);
BUG_ON(ret);
}
@@ -1118,8 +1176,11 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
wait_current_trans_commit_start_and_unblock(root, cur_trans);
else
wait_current_trans_commit_start(root, cur_trans);
- put_transaction(cur_trans);
+ if (current->journal_info == trans)
+ current->journal_info = NULL;
+
+ put_transaction(cur_trans);
return 0;
}
@@ -1238,21 +1299,42 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
schedule_timeout(1);
finish_wait(&cur_trans->writer_wait, &wait);
- spin_lock(&root->fs_info->trans_lock);
- root->fs_info->trans_no_join = 1;
- spin_unlock(&root->fs_info->trans_lock);
} while (atomic_read(&cur_trans->num_writers) > 1 ||
(should_grow && cur_trans->num_joined != joined));
- ret = create_pending_snapshots(trans, root->fs_info);
- BUG_ON(ret);
+ /*
+ * Ok now we need to make sure to block out any other joins while we
+ * commit the transaction. We could have started a join before setting
+ * no_join so make sure to wait for num_writers to == 1 again.
+ */
+ spin_lock(&root->fs_info->trans_lock);
+ root->fs_info->trans_no_join = 1;
+ spin_unlock(&root->fs_info->trans_lock);
+ wait_event(cur_trans->writer_wait,
+ atomic_read(&cur_trans->num_writers) == 1);
+
+ /*
+ * the reloc mutex makes sure that we stop
+ * the balancing code from coming in and moving
+ * extents around in the middle of the commit
+ */
+ mutex_lock(&root->fs_info->reloc_mutex);
ret = btrfs_run_delayed_items(trans, root);
BUG_ON(ret);
+ ret = create_pending_snapshots(trans, root->fs_info);
+ BUG_ON(ret);
+
ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
BUG_ON(ret);
+ /*
+ * make sure none of the code above managed to slip in a
+ * delayed item
+ */
+ btrfs_assert_delayed_root_empty(root);
+
WARN_ON(cur_trans != trans->transaction);
btrfs_scrub_pause(root);
@@ -1309,6 +1391,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
root->fs_info->running_transaction = NULL;
root->fs_info->trans_no_join = 0;
spin_unlock(&root->fs_info->trans_lock);
+ mutex_unlock(&root->fs_info->reloc_mutex);
wake_up(&root->fs_info->transaction_wait);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 592396c6dc47..4ce8a9f41d1e 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -3177,7 +3177,7 @@ again:
tmp_key.offset = (u64)-1;
wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key);
- BUG_ON(!wc.replay_dest);
+ BUG_ON(IS_ERR_OR_NULL(wc.replay_dest));
wc.replay_dest->log_root = log;
btrfs_record_root_in_trans(trans, wc.replay_dest);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index da541dfca2e3..1efa56e18f9b 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -689,12 +689,8 @@ int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
transid = btrfs_super_generation(disk_super);
if (disk_super->label[0])
printk(KERN_INFO "device label %s ", disk_super->label);
- else {
- /* FIXME, make a readl uuid parser */
- printk(KERN_INFO "device fsid %llx-%llx ",
- *(unsigned long long *)disk_super->fsid,
- *(unsigned long long *)(disk_super->fsid + 8));
- }
+ else
+ printk(KERN_INFO "device fsid %pU ", disk_super->fsid);
printk(KERN_CONT "devid %llu transid %llu %s\n",
(unsigned long long)devid, (unsigned long long)transid, path);
ret = device_list_add(path, disk_super, devid, fs_devices_ret);
diff --git a/fs/buffer.c b/fs/buffer.c
index 49c9aada0374..1a80b048ade8 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1902,10 +1902,8 @@ int __block_write_begin(struct page *page, loff_t pos, unsigned len,
if (!buffer_uptodate(*wait_bh))
err = -EIO;
}
- if (unlikely(err)) {
+ if (unlikely(err))
page_zero_new_buffers(page, from, to);
- ClearPageUptodate(page);
- }
return err;
}
EXPORT_SYMBOL(__block_write_begin);
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 33da49dc3cc6..5a3953db8118 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -453,7 +453,7 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc)
int err;
struct inode *inode = page->mapping->host;
BUG_ON(!inode);
- igrab(inode);
+ ihold(inode);
err = writepage_nounlock(page, wbc);
unlock_page(page);
iput(inode);
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 1f72b00447c4..f605753c8fe9 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -2940,14 +2940,12 @@ void ceph_flush_dirty_caps(struct ceph_mds_client *mdsc)
while (!list_empty(&mdsc->cap_dirty)) {
ci = list_first_entry(&mdsc->cap_dirty, struct ceph_inode_info,
i_dirty_item);
- inode = igrab(&ci->vfs_inode);
+ inode = &ci->vfs_inode;
+ ihold(inode);
dout("flush_dirty_caps %p\n", inode);
spin_unlock(&mdsc->cap_dirty_lock);
- if (inode) {
- ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH,
- NULL);
- iput(inode);
- }
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH, NULL);
+ iput(inode);
spin_lock(&mdsc->cap_dirty_lock);
}
spin_unlock(&mdsc->cap_dirty_lock);
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 33729e822bb9..ef8f08c343e8 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -308,7 +308,8 @@ more:
req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_dentry = dget(filp->f_dentry);
/* hints to request -> mds selection code */
req->r_direct_mode = USE_AUTH_MDS;
@@ -787,10 +788,12 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir,
req->r_dentry_drop = CEPH_CAP_FILE_SHARED;
req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
err = ceph_mdsc_do_request(mdsc, dir, req);
- if (err)
+ if (err) {
d_drop(dentry);
- else if (!req->r_reply_info.head->is_dentry)
- d_instantiate(dentry, igrab(old_dentry->d_inode));
+ } else if (!req->r_reply_info.head->is_dentry) {
+ ihold(old_dentry->d_inode);
+ d_instantiate(dentry, old_dentry->d_inode);
+ }
ceph_mdsc_put_request(req);
return err;
}
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index a610d3d67488..f67b687550de 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -109,7 +109,7 @@ static struct dentry *__fh_to_dentry(struct super_block *sb,
err = ceph_mdsc_do_request(mdsc, NULL, req);
inode = req->r_target_inode;
if (inode)
- igrab(inode);
+ ihold(inode);
ceph_mdsc_put_request(req);
if (!inode)
return ERR_PTR(-ESTALE);
@@ -167,7 +167,7 @@ static struct dentry *__cfh_to_dentry(struct super_block *sb,
err = ceph_mdsc_do_request(mdsc, NULL, req);
inode = req->r_target_inode;
if (inode)
- igrab(inode);
+ ihold(inode);
ceph_mdsc_put_request(req);
if (!inode)
return ERR_PTR(err ? err : -ESTALE);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 203252d88d9f..9542f07d0b93 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -191,7 +191,8 @@ int ceph_open(struct inode *inode, struct file *file)
err = PTR_ERR(req);
goto out;
}
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_num_caps = 1;
err = ceph_mdsc_do_request(mdsc, parent_inode, req);
if (!err)
@@ -282,7 +283,7 @@ int ceph_release(struct inode *inode, struct file *file)
static int striped_read(struct inode *inode,
u64 off, u64 len,
struct page **pages, int num_pages,
- int *checkeof, bool align_to_pages,
+ int *checkeof, bool o_direct,
unsigned long buf_align)
{
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
@@ -307,7 +308,7 @@ static int striped_read(struct inode *inode,
io_align = off & ~PAGE_MASK;
more:
- if (align_to_pages)
+ if (o_direct)
page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
else
page_align = pos & ~PAGE_MASK;
@@ -317,10 +318,10 @@ more:
ci->i_truncate_seq,
ci->i_truncate_size,
page_pos, pages_left, page_align);
- hit_stripe = this_len < left;
- was_short = ret >= 0 && ret < this_len;
if (ret == -ENOENT)
ret = 0;
+ hit_stripe = this_len < left;
+ was_short = ret >= 0 && ret < this_len;
dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read,
ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : "");
@@ -345,20 +346,22 @@ more:
}
if (was_short) {
- /* was original extent fully inside i_size? */
- if (pos + left <= inode->i_size) {
- dout("zero tail\n");
- ceph_zero_page_vector_range(page_off + read, len - read,
+ /* did we bounce off eof? */
+ if (pos + left > inode->i_size)
+ *checkeof = 1;
+
+ /* zero trailing bytes (inside i_size) */
+ if (left > 0 && pos < inode->i_size) {
+ if (pos + left > inode->i_size)
+ left = inode->i_size - pos;
+
+ dout("zero tail %d\n", left);
+ ceph_zero_page_vector_range(page_off + read, left,
pages);
- read = len;
- goto out;
+ read += left;
}
-
- /* check i_size */
- *checkeof = 1;
}
-out:
if (ret >= 0)
ret = read;
dout("striped_read returns %d\n", ret);
@@ -658,7 +661,7 @@ out:
/* hit EOF or hole? */
if (statret == 0 && *ppos < inode->i_size) {
- dout("aio_read sync_read hit hole, reading more\n");
+ dout("aio_read sync_read hit hole, ppos %lld < size %lld, reading more\n", *ppos, inode->i_size);
read += ret;
base += ret;
len -= ret;
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 70b6a4839c38..d8858e96ab18 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -1101,10 +1101,10 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
goto done;
}
req->r_dentry = dn; /* may have spliced */
- igrab(in);
+ ihold(in);
} else if (ceph_ino(in) == vino.ino &&
ceph_snap(in) == vino.snap) {
- igrab(in);
+ ihold(in);
} else {
dout(" %p links to %p %llx.%llx, not %llx.%llx\n",
dn, in, ceph_ino(in), ceph_snap(in),
@@ -1144,7 +1144,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
goto done;
}
req->r_dentry = dn; /* may have spliced */
- igrab(in);
+ ihold(in);
rinfo->head->is_dentry = 1; /* fool notrace handlers */
}
@@ -1328,7 +1328,7 @@ void ceph_queue_writeback(struct inode *inode)
if (queue_work(ceph_inode_to_client(inode)->wb_wq,
&ceph_inode(inode)->i_wb_work)) {
dout("ceph_queue_writeback %p\n", inode);
- igrab(inode);
+ ihold(inode);
} else {
dout("ceph_queue_writeback %p failed\n", inode);
}
@@ -1353,7 +1353,7 @@ void ceph_queue_invalidate(struct inode *inode)
if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq,
&ceph_inode(inode)->i_pg_inv_work)) {
dout("ceph_queue_invalidate %p\n", inode);
- igrab(inode);
+ ihold(inode);
} else {
dout("ceph_queue_invalidate %p failed\n", inode);
}
@@ -1477,7 +1477,7 @@ void ceph_queue_vmtruncate(struct inode *inode)
if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq,
&ci->i_vmtruncate_work)) {
dout("ceph_queue_vmtruncate %p\n", inode);
- igrab(inode);
+ ihold(inode);
} else {
dout("ceph_queue_vmtruncate %p failed, pending=%d\n",
inode, ci->i_truncate_pending);
@@ -1738,7 +1738,8 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
__mark_inode_dirty(inode, inode_dirty_flags);
if (mask) {
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = release;
req->r_args.setattr.mask = cpu_to_le32(mask);
req->r_num_caps = 1;
@@ -1779,7 +1780,8 @@ int ceph_do_getattr(struct inode *inode, int mask)
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_num_caps = 1;
req->r_args.getattr.mask = cpu_to_le32(mask);
err = ceph_mdsc_do_request(mdsc, NULL, req);
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 8888c9ba68db..ef0b5f48e13a 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -73,7 +73,8 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_EXCL;
req->r_args.setlayout.layout.fl_stripe_unit =
@@ -135,7 +136,8 @@ static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_args.setlayout.layout.fl_stripe_unit =
cpu_to_le32(l.stripe_unit);
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 476b329867d4..80576d05d687 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -23,7 +23,8 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
/* mds requires start and length rather than start and end */
if (LLONG_MAX == fl->fl_end)
@@ -32,11 +33,10 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
length = fl->fl_end - fl->fl_start + 1;
dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
- "length: %llu, wait: %d, type`: %d", (int)lock_type,
+ "length: %llu, wait: %d, type: %d", (int)lock_type,
(int)operation, (u64)fl->fl_pid, fl->fl_start,
length, wait, fl->fl_type);
-
req->r_args.filelock_change.rule = lock_type;
req->r_args.filelock_change.type = cmd;
req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid);
@@ -70,7 +70,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
}
ceph_mdsc_put_request(req);
dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
- "length: %llu, wait: %d, type`: %d, err code %d", (int)lock_type,
+ "length: %llu, wait: %d, type: %d, err code %d", (int)lock_type,
(int)operation, (u64)fl->fl_pid, fl->fl_start,
length, wait, fl->fl_type, err);
return err;
@@ -109,16 +109,20 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
dout("mds locked, locking locally");
err = posix_lock_file(file, fl, NULL);
if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
- /* undo! This should only happen if the kernel detects
- * local deadlock. */
+ /* undo! This should only happen if
+ * the kernel detects local
+ * deadlock. */
ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
CEPH_LOCK_UNLOCK, 0, fl);
- dout("got %d on posix_lock_file, undid lock", err);
+ dout("got %d on posix_lock_file, undid lock",
+ err);
}
}
- } else {
- dout("mds returned error code %d", err);
+ } else if (err == -ERESTARTSYS) {
+ dout("undoing lock\n");
+ ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
+ CEPH_LOCK_UNLOCK, 0, fl);
}
return err;
}
@@ -155,8 +159,11 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
file, CEPH_LOCK_UNLOCK, 0, fl);
dout("got %d on flock_lock_file_wait, undid lock", err);
}
- } else {
- dout("mds error code %d", err);
+ } else if (err == -ERESTARTSYS) {
+ dout("undoing lock\n");
+ ceph_lock_message(CEPH_LOCK_FLOCK,
+ CEPH_MDS_OP_SETFILELOCK,
+ file, CEPH_LOCK_UNLOCK, 0, fl);
}
return err;
}
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 24067d68a554..54b14de2e729 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -722,7 +722,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc)
ci = list_first_entry(&mdsc->snap_flush_list,
struct ceph_inode_info, i_snap_flush_item);
inode = &ci->vfs_inode;
- igrab(inode);
+ ihold(inode);
spin_unlock(&mdsc->snap_flush_lock);
spin_lock(&inode->i_lock);
__ceph_flush_snaps(ci, &session, 0);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index f2b628696180..f42d730f1b66 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -665,7 +665,8 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
err = PTR_ERR(req);
goto out;
}
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = CEPH_CAP_XATTR_SHARED;
req->r_num_caps = 1;
req->r_args.setxattr.flags = cpu_to_le32(flags);
@@ -795,7 +796,8 @@ static int ceph_send_removexattr(struct dentry *dentry, const char *name)
USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = CEPH_CAP_XATTR_SHARED;
req->r_num_caps = 1;
req->r_path2 = kstrdup(name, GFP_NOFS);
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index 1cd4c3a1862d..53ed1ad2c112 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -7,6 +7,7 @@ config CIFS
select CRYPTO_MD5
select CRYPTO_HMAC
select CRYPTO_ARC4
+ select CRYPTO_ECB
select CRYPTO_DES
help
This is the client VFS module for the Common Internet File System
@@ -148,7 +149,7 @@ config CIFS_FSCACHE
config CIFS_ACL
bool "Provide CIFS ACL support (EXPERIMENTAL)"
- depends on EXPERIMENTAL && CIFS_XATTR
+ depends on EXPERIMENTAL && CIFS_XATTR && KEYS
help
Allows to fetch CIFS/NTFS ACL from the server. The DACL blob
is handed over to the application/caller.
diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c
index dd8584d35a14..545509c3313b 100644
--- a/fs/cifs/cache.c
+++ b/fs/cifs/cache.c
@@ -92,7 +92,7 @@ static uint16_t cifs_server_get_key(const void *cookie_netfs_data,
break;
default:
- cERROR(1, "CIFS: Unknown network family '%d'", sa->sa_family);
+ cERROR(1, "Unknown network family '%d'", sa->sa_family);
key_len = 0;
break;
}
@@ -152,7 +152,7 @@ static uint16_t cifs_super_get_key(const void *cookie_netfs_data, void *buffer,
sharename = extract_sharename(tcon->treeName);
if (IS_ERR(sharename)) {
- cFYI(1, "CIFS: couldn't extract sharename\n");
+ cFYI(1, "%s: couldn't extract sharename\n", __func__);
sharename = NULL;
return 0;
}
@@ -302,7 +302,7 @@ static void cifs_fscache_inode_now_uncached(void *cookie_netfs_data)
pagevec_init(&pvec, 0);
first = 0;
- cFYI(1, "cifs inode 0x%p now uncached", cifsi);
+ cFYI(1, "%s: cifs inode 0x%p now uncached", __func__, cifsi);
for (;;) {
nr_pages = pagevec_lookup(&pvec,
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index dfbd9f1f373d..5a0ee7f2af06 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -184,7 +184,7 @@ int cifs_verify_signature(struct smb_hdr *cifs_pdu,
if (cifs_pdu == NULL || server == NULL)
return -EINVAL;
- if (cifs_pdu->Command == SMB_COM_NEGOTIATE)
+ if (!server->session_estab)
return 0;
if (cifs_pdu->Command == SMB_COM_LOCKING_ANDX) {
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 989442dcfb45..2f0c58646c10 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -257,9 +257,6 @@ static int cifs_permission(struct inode *inode, int mask, unsigned int flags)
{
struct cifs_sb_info *cifs_sb;
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
-
cifs_sb = CIFS_SB(inode->i_sb);
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) {
@@ -352,6 +349,37 @@ cifs_show_address(struct seq_file *s, struct TCP_Server_Info *server)
}
}
+static void
+cifs_show_security(struct seq_file *s, struct TCP_Server_Info *server)
+{
+ seq_printf(s, ",sec=");
+
+ switch (server->secType) {
+ case LANMAN:
+ seq_printf(s, "lanman");
+ break;
+ case NTLMv2:
+ seq_printf(s, "ntlmv2");
+ break;
+ case NTLM:
+ seq_printf(s, "ntlm");
+ break;
+ case Kerberos:
+ seq_printf(s, "krb5");
+ break;
+ case RawNTLMSSP:
+ seq_printf(s, "ntlmssp");
+ break;
+ default:
+ /* shouldn't ever happen */
+ seq_printf(s, "unknown");
+ break;
+ }
+
+ if (server->sec_mode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED))
+ seq_printf(s, "i");
+}
+
/*
* cifs_show_options() is for displaying mount options in /proc/mounts.
* Not all settable options are displayed but most of the important
@@ -365,6 +393,8 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
struct sockaddr *srcaddr;
srcaddr = (struct sockaddr *)&tcon->ses->server->srcaddr;
+ cifs_show_security(s, tcon->ses->server);
+
seq_printf(s, ",unc=%s", tcon->treeName);
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 64313f778ebf..0900e1658c96 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -129,5 +129,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
extern const struct export_operations cifs_export_ops;
#endif /* CIFS_NFSD_EXPORT */
-#define CIFS_VERSION "1.72"
+#define CIFS_VERSION "1.73"
#endif /* _CIFSFS_H */
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 6d88b82537c3..12cf72dd0c42 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -152,7 +152,7 @@ cifs_reconnect(struct TCP_Server_Info *server)
mid_entry->callback(mid_entry);
}
- while (server->tcpStatus == CifsNeedReconnect) {
+ do {
try_to_freeze();
/* we should try only the port we connected to before */
@@ -167,7 +167,7 @@ cifs_reconnect(struct TCP_Server_Info *server)
server->tcpStatus = CifsNeedNegotiate;
spin_unlock(&GlobalMid_Lock);
}
- }
+ } while (server->tcpStatus == CifsNeedReconnect);
return rc;
}
@@ -784,7 +784,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
struct smb_vol *vol)
{
char *value, *data, *end;
- char *mountdata_copy, *options;
+ char *mountdata_copy = NULL, *options;
unsigned int temp_len, i, j;
char separator[2];
short int override_uid = -1;
@@ -1391,7 +1391,7 @@ cifs_parse_mount_options(const char *mountdata, const char *devname,
"/proc/fs/cifs/LookupCacheEnabled to 0\n");
} else if (strnicmp(data, "fsc", 3) == 0) {
#ifndef CONFIG_CIFS_FSCACHE
- cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE"
+ cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE "
"kernel config option set");
goto cifs_parse_mount_err;
#endif
@@ -1976,7 +1976,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info)
warned_on_ntlm = true;
cERROR(1, "default security mechanism requested. The default "
"security mechanism will be upgraded from ntlm to "
- "ntlmv2 in kernel release 2.6.41");
+ "ntlmv2 in kernel release 3.1");
}
ses->overrideSecFlg = volume_info->secFlg;
@@ -2149,7 +2149,10 @@ cifs_put_tlink(struct tcon_link *tlink)
}
static inline struct tcon_link *
-cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb);
+cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb)
+{
+ return cifs_sb->master_tlink;
+}
static int
compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data)
@@ -3171,6 +3174,10 @@ out:
return rc;
}
+/*
+ * Issue a TREE_CONNECT request. Note that for IPC$ shares, that the tcon
+ * pointer may be NULL.
+ */
int
CIFSTCon(unsigned int xid, struct cifs_ses *ses,
const char *tree, struct cifs_tcon *tcon,
@@ -3205,7 +3212,7 @@ CIFSTCon(unsigned int xid, struct cifs_ses *ses,
pSMB->AndXCommand = 0xFF;
pSMB->Flags = cpu_to_le16(TCON_EXTENDED_SECINFO);
bcc_ptr = &pSMB->Password[0];
- if ((ses->server->sec_mode) & SECMODE_USER) {
+ if (!tcon || (ses->server->sec_mode & SECMODE_USER)) {
pSMB->PasswordLength = cpu_to_le16(1); /* minimum */
*bcc_ptr = 0; /* password is null byte */
bcc_ptr++; /* skip password */
@@ -3371,7 +3378,7 @@ int cifs_negotiate_protocol(unsigned int xid, struct cifs_ses *ses)
}
if (rc == 0) {
spin_lock(&GlobalMid_Lock);
- if (server->tcpStatus != CifsExiting)
+ if (server->tcpStatus == CifsNeedNegotiate)
server->tcpStatus = CifsGood;
else
rc = -EHOSTDOWN;
@@ -3484,12 +3491,6 @@ out:
return tcon;
}
-static inline struct tcon_link *
-cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb)
-{
- return cifs_sb->master_tlink;
-}
-
struct cifs_tcon *
cifs_sb_master_tcon(struct cifs_sb_info *cifs_sb)
{
diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c
index d368a47ba5eb..816696621ec9 100644
--- a/fs/cifs/fscache.c
+++ b/fs/cifs/fscache.c
@@ -28,14 +28,14 @@ void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server)
server->fscache =
fscache_acquire_cookie(cifs_fscache_netfs.primary_index,
&cifs_fscache_server_index_def, server);
- cFYI(1, "CIFS: get client cookie (0x%p/0x%p)", server,
- server->fscache);
+ cFYI(1, "%s: (0x%p/0x%p)", __func__, server,
+ server->fscache);
}
void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server)
{
- cFYI(1, "CIFS: release client cookie (0x%p/0x%p)", server,
- server->fscache);
+ cFYI(1, "%s: (0x%p/0x%p)", __func__, server,
+ server->fscache);
fscache_relinquish_cookie(server->fscache, 0);
server->fscache = NULL;
}
@@ -47,13 +47,13 @@ void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
tcon->fscache =
fscache_acquire_cookie(server->fscache,
&cifs_fscache_super_index_def, tcon);
- cFYI(1, "CIFS: get superblock cookie (0x%p/0x%p)",
- server->fscache, tcon->fscache);
+ cFYI(1, "%s: (0x%p/0x%p)", __func__, server->fscache,
+ tcon->fscache);
}
void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon)
{
- cFYI(1, "CIFS: releasing superblock cookie (0x%p)", tcon->fscache);
+ cFYI(1, "%s: (0x%p)", __func__, tcon->fscache);
fscache_relinquish_cookie(tcon->fscache, 0);
tcon->fscache = NULL;
}
@@ -70,8 +70,8 @@ static void cifs_fscache_enable_inode_cookie(struct inode *inode)
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) {
cifsi->fscache = fscache_acquire_cookie(tcon->fscache,
&cifs_fscache_inode_object_def, cifsi);
- cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache,
- cifsi->fscache);
+ cFYI(1, "%s: got FH cookie (0x%p/0x%p)", __func__,
+ tcon->fscache, cifsi->fscache);
}
}
@@ -80,8 +80,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode)
struct cifsInodeInfo *cifsi = CIFS_I(inode);
if (cifsi->fscache) {
- cFYI(1, "CIFS releasing inode cookie (0x%p)",
- cifsi->fscache);
+ cFYI(1, "%s: (0x%p)", __func__, cifsi->fscache);
fscache_relinquish_cookie(cifsi->fscache, 0);
cifsi->fscache = NULL;
}
@@ -92,8 +91,7 @@ static void cifs_fscache_disable_inode_cookie(struct inode *inode)
struct cifsInodeInfo *cifsi = CIFS_I(inode);
if (cifsi->fscache) {
- cFYI(1, "CIFS disabling inode cookie (0x%p)",
- cifsi->fscache);
+ cFYI(1, "%s: (0x%p)", __func__, cifsi->fscache);
fscache_relinquish_cookie(cifsi->fscache, 1);
cifsi->fscache = NULL;
}
@@ -121,8 +119,8 @@ void cifs_fscache_reset_inode_cookie(struct inode *inode)
cifs_sb_master_tcon(cifs_sb)->fscache,
&cifs_fscache_inode_object_def,
cifsi);
- cFYI(1, "CIFS: new cookie 0x%p oldcookie 0x%p",
- cifsi->fscache, old);
+ cFYI(1, "%s: new cookie 0x%p oldcookie 0x%p",
+ __func__, cifsi->fscache, old);
}
}
@@ -132,8 +130,8 @@ int cifs_fscache_release_page(struct page *page, gfp_t gfp)
struct inode *inode = page->mapping->host;
struct cifsInodeInfo *cifsi = CIFS_I(inode);
- cFYI(1, "CIFS: fscache release page (0x%p/0x%p)",
- page, cifsi->fscache);
+ cFYI(1, "%s: (0x%p/0x%p)", __func__, page,
+ cifsi->fscache);
if (!fscache_maybe_release_page(cifsi->fscache, page, gfp))
return 0;
}
@@ -144,8 +142,7 @@ int cifs_fscache_release_page(struct page *page, gfp_t gfp)
static void cifs_readpage_from_fscache_complete(struct page *page, void *ctx,
int error)
{
- cFYI(1, "CFS: readpage_from_fscache_complete (0x%p/%d)",
- page, error);
+ cFYI(1, "%s: (0x%p/%d)", __func__, page, error);
if (!error)
SetPageUptodate(page);
unlock_page(page);
@@ -158,7 +155,7 @@ int __cifs_readpage_from_fscache(struct inode *inode, struct page *page)
{
int ret;
- cFYI(1, "CIFS: readpage_from_fscache(fsc:%p, p:%p, i:0x%p",
+ cFYI(1, "%s: (fsc:%p, p:%p, i:0x%p", __func__,
CIFS_I(inode)->fscache, page, inode);
ret = fscache_read_or_alloc_page(CIFS_I(inode)->fscache, page,
cifs_readpage_from_fscache_complete,
@@ -167,11 +164,11 @@ int __cifs_readpage_from_fscache(struct inode *inode, struct page *page)
switch (ret) {
case 0: /* page found in fscache, read submitted */
- cFYI(1, "CIFS: readpage_from_fscache: submitted");
+ cFYI(1, "%s: submitted", __func__);
return ret;
case -ENOBUFS: /* page won't be cached */
case -ENODATA: /* page not in cache */
- cFYI(1, "CIFS: readpage_from_fscache %d", ret);
+ cFYI(1, "%s: %d", __func__, ret);
return 1;
default:
@@ -190,7 +187,7 @@ int __cifs_readpages_from_fscache(struct inode *inode,
{
int ret;
- cFYI(1, "CIFS: __cifs_readpages_from_fscache (0x%p/%u/0x%p)",
+ cFYI(1, "%s: (0x%p/%u/0x%p)", __func__,
CIFS_I(inode)->fscache, *nr_pages, inode);
ret = fscache_read_or_alloc_pages(CIFS_I(inode)->fscache, mapping,
pages, nr_pages,
@@ -199,12 +196,12 @@ int __cifs_readpages_from_fscache(struct inode *inode,
mapping_gfp_mask(mapping));
switch (ret) {
case 0: /* read submitted to the cache for all pages */
- cFYI(1, "CIFS: readpages_from_fscache: submitted");
+ cFYI(1, "%s: submitted", __func__);
return ret;
case -ENOBUFS: /* some pages are not cached and can't be */
case -ENODATA: /* some pages are not cached */
- cFYI(1, "CIFS: readpages_from_fscache: no page");
+ cFYI(1, "%s: no page", __func__);
return 1;
default:
@@ -218,7 +215,7 @@ void __cifs_readpage_to_fscache(struct inode *inode, struct page *page)
{
int ret;
- cFYI(1, "CIFS: readpage_to_fscache(fsc: %p, p: %p, i: %p",
+ cFYI(1, "%s: (fsc: %p, p: %p, i: %p)", __func__,
CIFS_I(inode)->fscache, page, inode);
ret = fscache_write_page(CIFS_I(inode)->fscache, page, GFP_KERNEL);
if (ret != 0)
@@ -230,7 +227,7 @@ void __cifs_fscache_invalidate_page(struct page *page, struct inode *inode)
struct cifsInodeInfo *cifsi = CIFS_I(inode);
struct fscache_cookie *cookie = cifsi->fscache;
- cFYI(1, "CIFS: fscache invalidatepage (0x%p/0x%p)", page, cookie);
+ cFYI(1, "%s: (0x%p/0x%p)", __func__, page, cookie);
fscache_wait_on_page_write(cookie, page);
fscache_uncache_page(cookie, page);
}
diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c
index 6cbb3afb36dc..cb140ef293e4 100644
--- a/fs/coda/pioctl.c
+++ b/fs/coda/pioctl.c
@@ -43,8 +43,6 @@ const struct file_operations coda_ioctl_operations = {
/* the coda pioctl inode ops */
static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags)
{
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
return (mask & MAY_EXEC) ? -EACCES : 0;
}
diff --git a/fs/dcookies.c b/fs/dcookies.c
index a21cabdbd87b..dda0dc702d1b 100644
--- a/fs/dcookies.c
+++ b/fs/dcookies.c
@@ -178,6 +178,8 @@ SYSCALL_DEFINE(lookup_dcookie)(u64 cookie64, char __user * buf, size_t len)
/* FIXME: (deleted) ? */
path = d_path(&dcs->path, kbuf, PAGE_SIZE);
+ mutex_unlock(&dcookie_mutex);
+
if (IS_ERR(path)) {
err = PTR_ERR(path);
goto out_free;
@@ -194,6 +196,7 @@ SYSCALL_DEFINE(lookup_dcookie)(u64 cookie64, char __user * buf, size_t len)
out_free:
kfree(kbuf);
+ return err;
out:
mutex_unlock(&dcookie_mutex);
return err;
diff --git a/fs/exec.c b/fs/exec.c
index ea5f748906a8..6075a1e727ae 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1093,6 +1093,7 @@ int flush_old_exec(struct linux_binprm * bprm)
bprm->mm = NULL; /* We're using it now */
+ set_fs(USER_DS);
current->flags &= ~(PF_RANDOMIZE | PF_KTHREAD);
flush_thread();
current->personality &= ~bprm->per_clear;
@@ -1357,10 +1358,6 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
if (retval)
return retval;
- /* kernel module loader fixup */
- /* so we don't try to load run modprobe in kernel space. */
- set_fs(USER_DS);
-
retval = audit_bprm(bprm);
if (retval)
return retval;
@@ -1999,7 +1996,7 @@ static void wait_for_dump_helpers(struct file *file)
* is a special value that we use to trap recursive
* core dumps
*/
-static int umh_pipe_setup(struct subprocess_info *info)
+static int umh_pipe_setup(struct subprocess_info *info, struct cred *new)
{
struct file *rp, *wp;
struct fdtable *fdt;
diff --git a/fs/fat/file.c b/fs/fat/file.c
index 7257752b6d5d..7018e1d8902d 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -102,7 +102,7 @@ static int fat_ioctl_set_attributes(struct file *file, u32 __user *user_attr)
if (attr & ATTR_SYS)
inode->i_flags |= S_IMMUTABLE;
else
- inode->i_flags &= S_IMMUTABLE;
+ inode->i_flags &= ~S_IMMUTABLE;
}
fat_save_attrs(inode, attr);
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index cc6ec4b2f0ff..38f84cd48b67 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -921,6 +921,8 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
if (sb->s_flags & MS_MANDLOCK)
goto err;
+ sb->s_flags &= ~MS_NOSEC;
+
if (!parse_fuse_opt((char *) data, &d, is_bdev))
goto err;
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index 2792a790e50b..1c1336e7b3b2 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -663,14 +663,19 @@ static void glock_work_func(struct work_struct *work)
drop_ref = 1;
}
spin_lock(&gl->gl_spin);
- if (test_and_clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
+ if (test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) &&
gl->gl_state != LM_ST_UNLOCKED &&
gl->gl_demote_state != LM_ST_EXCLUSIVE) {
unsigned long holdtime, now = jiffies;
+
holdtime = gl->gl_tchange + gl->gl_ops->go_min_hold_time;
if (time_before(now, holdtime))
delay = holdtime - now;
- set_bit(delay ? GLF_PENDING_DEMOTE : GLF_DEMOTE, &gl->gl_flags);
+
+ if (!delay) {
+ clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags);
+ set_bit(GLF_DEMOTE, &gl->gl_flags);
+ }
}
run_queue(gl, 0);
spin_unlock(&gl->gl_spin);
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 3db5ba4568fc..b3cc8586984e 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -974,7 +974,7 @@ out_no_inode:
out_no_read:
printk(KERN_WARNING "%s: bread failed, dev=%s, iso_blknum=%d, block=%d\n",
__func__, s->s_id, iso_blknum, block);
- goto out_freesbi;
+ goto out_freebh;
out_bad_zone_size:
printk(KERN_WARNING "ISOFS: Bad logical zone size %ld\n",
sbi->s_log_zone_size);
@@ -989,6 +989,7 @@ out_unknown_format:
out_freebh:
brelse(bh);
+ brelse(pri_bh);
out_freesbi:
kfree(opt.iocharset);
kfree(sbi);
diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c
index 278e3fb40b71..583636f745e5 100644
--- a/fs/jfs/jfs_logmgr.c
+++ b/fs/jfs/jfs_logmgr.c
@@ -1123,7 +1123,7 @@ int lmLogOpen(struct super_block *sb)
bdev = blkdev_get_by_dev(sbi->logdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL,
log);
if (IS_ERR(bdev)) {
- rc = -PTR_ERR(bdev);
+ rc = PTR_ERR(bdev);
goto free;
}
diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c
index 9ed89d1663f8..1afae26cf236 100644
--- a/fs/logfs/dir.c
+++ b/fs/logfs/dir.c
@@ -555,13 +555,6 @@ static int logfs_symlink(struct inode *dir, struct dentry *dentry,
return __logfs_create(dir, dentry, inode, target, destlen);
}
-static int logfs_permission(struct inode *inode, int mask, unsigned int flags)
-{
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
- return generic_permission(inode, mask, flags, NULL);
-}
-
static int logfs_link(struct dentry *old_dentry, struct inode *dir,
struct dentry *dentry)
{
@@ -820,7 +813,6 @@ const struct inode_operations logfs_dir_iops = {
.mknod = logfs_mknod,
.rename = logfs_rename,
.rmdir = logfs_rmdir,
- .permission = logfs_permission,
.symlink = logfs_symlink,
.unlink = logfs_unlink,
};
diff --git a/fs/namei.c b/fs/namei.c
index e2e4e8d032ee..0223c41fb114 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -238,7 +238,8 @@ int generic_permission(struct inode *inode, int mask, unsigned int flags,
/*
* Read/write DACs are always overridable.
- * Executable DACs are overridable if at least one exec bit is set.
+ * Executable DACs are overridable for all directories and
+ * for non-directories that have least one exec bit set.
*/
if (!(mask & MAY_EXEC) || execute_ok(inode))
if (ns_capable(inode_userns(inode), CAP_DAC_OVERRIDE))
@@ -812,6 +813,11 @@ static int follow_automount(struct path *path, unsigned flags,
if (!mnt) /* mount collision */
return 0;
+ if (!*need_mntput) {
+ /* lock_mount() may release path->mnt on error */
+ mntget(path->mnt);
+ *need_mntput = true;
+ }
err = finish_automount(mnt, path);
switch (err) {
@@ -819,12 +825,9 @@ static int follow_automount(struct path *path, unsigned flags,
/* Someone else made a mount here whilst we were busy */
return 0;
case 0:
- dput(path->dentry);
- if (*need_mntput)
- mntput(path->mnt);
+ path_put(path);
path->mnt = mnt;
path->dentry = dget(mnt->mnt_root);
- *need_mntput = true;
return 0;
default:
return err;
@@ -844,9 +847,10 @@ static int follow_automount(struct path *path, unsigned flags,
*/
static int follow_managed(struct path *path, unsigned flags)
{
+ struct vfsmount *mnt = path->mnt; /* held by caller, must be left alone */
unsigned managed;
bool need_mntput = false;
- int ret;
+ int ret = 0;
/* Given that we're not holding a lock here, we retain the value in a
* local variable for each dentry as we look at it so that we don't see
@@ -861,7 +865,7 @@ static int follow_managed(struct path *path, unsigned flags)
BUG_ON(!path->dentry->d_op->d_manage);
ret = path->dentry->d_op->d_manage(path->dentry, false);
if (ret < 0)
- return ret == -EISDIR ? 0 : ret;
+ break;
}
/* Transit to a mounted filesystem. */
@@ -887,14 +891,19 @@ static int follow_managed(struct path *path, unsigned flags)
if (managed & DCACHE_NEED_AUTOMOUNT) {
ret = follow_automount(path, flags, &need_mntput);
if (ret < 0)
- return ret == -EISDIR ? 0 : ret;
+ break;
continue;
}
/* We didn't change the current path point */
break;
}
- return 0;
+
+ if (need_mntput && path->mnt == mnt)
+ mntput(path->mnt);
+ if (ret == -EISDIR)
+ ret = 0;
+ return ret;
}
int follow_down_one(struct path *path)
@@ -1003,9 +1012,6 @@ failed:
* Follow down to the covering mount currently visible to userspace. At each
* point, the filesystem owning that dentry may be queried as to whether the
* caller is permitted to proceed or not.
- *
- * Care must be taken as namespace_sem may be held (indicated by mounting_here
- * being true).
*/
int follow_down(struct path *path)
{
@@ -2624,6 +2630,10 @@ static long do_rmdir(int dfd, const char __user *pathname)
error = PTR_ERR(dentry);
if (IS_ERR(dentry))
goto exit2;
+ if (!dentry->d_inode) {
+ error = -ENOENT;
+ goto exit3;
+ }
error = mnt_want_write(nd.path.mnt);
if (error)
goto exit3;
@@ -2712,8 +2722,9 @@ static long do_unlinkat(int dfd, const char __user *pathname)
if (nd.last.name[nd.last.len])
goto slashes;
inode = dentry->d_inode;
- if (inode)
- ihold(inode);
+ if (!inode)
+ goto slashes;
+ ihold(inode);
error = mnt_want_write(nd.path.mnt);
if (error)
goto exit2;
diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c
index 7eafe468a29c..b2e3ff347620 100644
--- a/fs/nilfs2/btree.c
+++ b/fs/nilfs2/btree.c
@@ -1346,6 +1346,11 @@ static void nilfs_btree_shrink(struct nilfs_bmap *btree,
path[level].bp_bh = NULL;
}
+static void nilfs_btree_nop(struct nilfs_bmap *btree,
+ struct nilfs_btree_path *path,
+ int level, __u64 *keyp, __u64 *ptrp)
+{
+}
static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree,
struct nilfs_btree_path *path,
@@ -1356,20 +1361,19 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree,
struct buffer_head *bh;
struct nilfs_btree_node *node, *parent, *sib;
__u64 sibptr;
- int pindex, level, ncmin, ncmax, ncblk, ret;
+ int pindex, dindex, level, ncmin, ncmax, ncblk, ret;
ret = 0;
stats->bs_nblocks = 0;
ncmin = NILFS_BTREE_NODE_NCHILDREN_MIN(nilfs_btree_node_size(btree));
ncblk = nilfs_btree_nchildren_per_block(btree);
- for (level = NILFS_BTREE_LEVEL_NODE_MIN;
+ for (level = NILFS_BTREE_LEVEL_NODE_MIN, dindex = path[level].bp_index;
level < nilfs_btree_height(btree) - 1;
level++) {
node = nilfs_btree_get_nonroot_node(path, level);
path[level].bp_oldreq.bpr_ptr =
- nilfs_btree_node_get_ptr(node, path[level].bp_index,
- ncblk);
+ nilfs_btree_node_get_ptr(node, dindex, ncblk);
ret = nilfs_bmap_prepare_end_ptr(btree,
&path[level].bp_oldreq, dat);
if (ret < 0)
@@ -1383,6 +1387,7 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree,
parent = nilfs_btree_get_node(btree, path, level + 1, &ncmax);
pindex = path[level + 1].bp_index;
+ dindex = pindex;
if (pindex > 0) {
/* left sibling */
@@ -1421,6 +1426,14 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree,
path[level].bp_sib_bh = bh;
path[level].bp_op = nilfs_btree_concat_right;
stats->bs_nblocks++;
+ /*
+ * When merging right sibling node
+ * into the current node, pointer to
+ * the right sibling node must be
+ * terminated instead. The adjustment
+ * below is required for that.
+ */
+ dindex = pindex + 1;
/* continue; */
}
} else {
@@ -1431,29 +1444,31 @@ static int nilfs_btree_prepare_delete(struct nilfs_bmap *btree,
NILFS_BTREE_ROOT_NCHILDREN_MAX) {
path[level].bp_op = nilfs_btree_shrink;
stats->bs_nblocks += 2;
+ level++;
+ path[level].bp_op = nilfs_btree_nop;
+ goto shrink_root_child;
} else {
path[level].bp_op = nilfs_btree_do_delete;
stats->bs_nblocks++;
+ goto out;
}
-
- goto out;
-
}
}
+ /* child of the root node is deleted */
+ path[level].bp_op = nilfs_btree_do_delete;
+ stats->bs_nblocks++;
+
+shrink_root_child:
node = nilfs_btree_get_root(btree);
path[level].bp_oldreq.bpr_ptr =
- nilfs_btree_node_get_ptr(node, path[level].bp_index,
+ nilfs_btree_node_get_ptr(node, dindex,
NILFS_BTREE_ROOT_NCHILDREN_MAX);
ret = nilfs_bmap_prepare_end_ptr(btree, &path[level].bp_oldreq, dat);
if (ret < 0)
goto err_out_child_node;
- /* child of the root node is deleted */
- path[level].bp_op = nilfs_btree_do_delete;
- stats->bs_nblocks++;
-
/* success */
out:
*levelp = level;
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index b954878ad6ce..b9b45fc2903e 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -801,12 +801,7 @@ out_err:
int nilfs_permission(struct inode *inode, int mask, unsigned int flags)
{
- struct nilfs_root *root;
-
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
-
- root = NILFS_I(inode)->i_root;
+ struct nilfs_root *root = NILFS_I(inode)->i_root;
if ((mask & MAY_WRITE) && root &&
root->cno != NILFS_CPTREE_CURRENT_CNO)
return -EROFS; /* snapshot is not writable */
diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 141646e88fb5..bb24ab6c282f 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -2573,7 +2573,7 @@ static struct nilfs_sc_info *nilfs_segctor_new(struct super_block *sb,
sci->sc_watermark = NILFS_SC_DEFAULT_WATERMARK;
if (nilfs->ns_interval)
- sci->sc_interval = nilfs->ns_interval;
+ sci->sc_interval = HZ * nilfs->ns_interval;
if (nilfs->ns_watermark)
sci->sc_watermark = nilfs->ns_watermark;
return sci;
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index cdbaf5e97308..56f61027236b 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1072,7 +1072,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
sb->s_magic = OCFS2_SUPER_MAGIC;
- sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+ sb->s_flags = (sb->s_flags & ~(MS_POSIXACL | MS_NOSEC)) |
((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
/* Hard readonly mode only if: bdev_read_only, MS_RDONLY,
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 14def991d9dd..8a84210ca080 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2169,11 +2169,7 @@ static const struct file_operations proc_fd_operations = {
*/
static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags)
{
- int rv;
-
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
- rv = generic_permission(inode, mask, flags, NULL);
+ int rv = generic_permission(inode, mask, flags, NULL);
if (rv == 0)
return 0;
if (task_pid(current) == proc_pid(inode))
diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c
index 781dec5bd682..be177f702acb 100644
--- a/fs/proc/namespaces.c
+++ b/fs/proc/namespaces.c
@@ -38,18 +38,21 @@ static struct dentry *proc_ns_instantiate(struct inode *dir,
struct inode *inode;
struct proc_inode *ei;
struct dentry *error = ERR_PTR(-ENOENT);
+ void *ns;
inode = proc_pid_make_inode(dir->i_sb, task);
if (!inode)
goto out;
+ ns = ns_ops->get(task);
+ if (!ns)
+ goto out_iput;
+
ei = PROC_I(inode);
inode->i_mode = S_IFREG|S_IRUSR;
inode->i_fop = &ns_file_operations;
ei->ns_ops = ns_ops;
- ei->ns = ns_ops->get(task);
- if (!ei->ns)
- goto out_iput;
+ ei->ns = ns;
dentry->d_op = &pid_dentry_operations;
d_add(dentry, inode);
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index f50133c11c24..d167de365a8d 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -304,9 +304,6 @@ static int proc_sys_permission(struct inode *inode, int mask,unsigned int flags)
struct ctl_table *table;
int error;
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
-
/* Executable files are not allowed under /proc/sys/ */
if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode))
return -EACCES;
diff --git a/fs/proc/root.c b/fs/proc/root.c
index a9000e9cfee5..d6c3b416529b 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -28,11 +28,12 @@ static int proc_test_super(struct super_block *sb, void *data)
static int proc_set_super(struct super_block *sb, void *data)
{
- struct pid_namespace *ns;
-
- ns = (struct pid_namespace *)data;
- sb->s_fs_info = get_pid_ns(ns);
- return set_anon_super(sb, NULL);
+ int err = set_anon_super(sb, NULL);
+ if (!err) {
+ struct pid_namespace *ns = (struct pid_namespace *)data;
+ sb->s_fs_info = get_pid_ns(ns);
+ }
+ return err;
}
static struct dentry *proc_mount(struct file_system_type *fs_type,
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index e8a62f41b458..d78089690965 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -954,8 +954,6 @@ static int xattr_mount_check(struct super_block *s)
int reiserfs_permission(struct inode *inode, int mask, unsigned int flags)
{
- if (flags & IPERM_FLAG_RCU)
- return -ECHILD;
/*
* We don't do permission checks on the internal objects.
* Permissions are determined by the "owning" object.
diff --git a/fs/super.c b/fs/super.c
index c75593953c52..ab3d672db0de 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -822,7 +822,7 @@ struct dentry *mount_bdev(struct file_system_type *fs_type,
} else {
char b[BDEVNAME_SIZE];
- s->s_flags = flags;
+ s->s_flags = flags | MS_NOSEC;
s->s_mode = mode;
strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id));
sb_set_blocksize(s, block_size(bdev));
diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c
index 266895783b47..e34f0d99ea4e 100644
--- a/fs/sysfs/mount.c
+++ b/fs/sysfs/mount.c
@@ -95,6 +95,14 @@ static int sysfs_set_super(struct super_block *sb, void *data)
return error;
}
+static void free_sysfs_super_info(struct sysfs_super_info *info)
+{
+ int type;
+ for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++)
+ kobj_ns_drop(type, info->ns[type]);
+ kfree(info);
+}
+
static struct dentry *sysfs_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
@@ -108,11 +116,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type,
return ERR_PTR(-ENOMEM);
for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++)
- info->ns[type] = kobj_ns_current(type);
+ info->ns[type] = kobj_ns_grab_current(type);
sb = sget(fs_type, sysfs_test_super, sysfs_set_super, info);
if (IS_ERR(sb) || sb->s_fs_info != info)
- kfree(info);
+ free_sysfs_super_info(info);
if (IS_ERR(sb))
return ERR_CAST(sb);
if (!sb->s_root) {
@@ -131,12 +139,11 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type,
static void sysfs_kill_sb(struct super_block *sb)
{
struct sysfs_super_info *info = sysfs_info(sb);
-
/* Remove the superblock from fs_supers/s_instances
* so we can't find it, before freeing sysfs_super_info.
*/
kill_anon_super(sb);
- kfree(info);
+ free_sysfs_super_info(info);
}
static struct file_system_type sysfs_fs_type = {
@@ -145,28 +152,6 @@ static struct file_system_type sysfs_fs_type = {
.kill_sb = sysfs_kill_sb,
};
-void sysfs_exit_ns(enum kobj_ns_type type, const void *ns)
-{
- struct super_block *sb;
-
- mutex_lock(&sysfs_mutex);
- spin_lock(&sb_lock);
- list_for_each_entry(sb, &sysfs_fs_type.fs_supers, s_instances) {
- struct sysfs_super_info *info = sysfs_info(sb);
- /*
- * If we see a superblock on the fs_supers/s_instances
- * list the unmount has not completed and sb->s_fs_info
- * points to a valid struct sysfs_super_info.
- */
- /* Ignore superblocks with the wrong ns */
- if (info->ns[type] != ns)
- continue;
- info->ns[type] = NULL;
- }
- spin_unlock(&sb_lock);
- mutex_unlock(&sysfs_mutex);
-}
-
int __init sysfs_init(void)
{
int err = -ENOMEM;
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 3d28af31d863..2ed2404f3113 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -136,7 +136,7 @@ struct sysfs_addrm_cxt {
* instance).
*/
struct sysfs_super_info {
- const void *ns[KOBJ_NS_TYPES];
+ void *ns[KOBJ_NS_TYPES];
};
#define sysfs_info(SB) ((struct sysfs_super_info *)(SB->s_fs_info))
extern struct sysfs_dirent sysfs_root;
diff --git a/fs/timerfd.c b/fs/timerfd.c
index f67acbdda5e8..dffeb3795af1 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -61,7 +61,9 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr)
/*
* Called when the clock was set to cancel the timers in the cancel
- * list.
+ * list. This will wake up processes waiting on these timers. The
+ * wake-up requires ctx->ticks to be non zero, therefore we increment
+ * it before calling wake_up_locked().
*/
void timerfd_clock_was_set(void)
{
@@ -76,6 +78,7 @@ void timerfd_clock_was_set(void)
spin_lock_irqsave(&ctx->wqh.lock, flags);
if (ctx->moffs.tv64 != moffs.tv64) {
ctx->moffs.tv64 = KTIME_MAX;
+ ctx->ticks++;
wake_up_locked(&ctx->wqh);
}
spin_unlock_irqrestore(&ctx->wqh.lock, flags);
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index b5aeb5a8ebed..529be0582029 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1848,7 +1848,6 @@ static void ubifs_put_super(struct super_block *sb)
bdi_destroy(&c->bdi);
ubi_close_volume(c->ubi);
mutex_unlock(&c->umount_mutex);
- kfree(c);
}
static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data)
@@ -1971,61 +1970,65 @@ static struct ubi_volume_desc *open_ubi(const char *name, int mode)
return ERR_PTR(-EINVAL);
}
-static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
+static struct ubifs_info *alloc_ubifs_info(struct ubi_volume_desc *ubi)
{
- struct ubi_volume_desc *ubi = sb->s_fs_info;
struct ubifs_info *c;
- struct inode *root;
- int err;
c = kzalloc(sizeof(struct ubifs_info), GFP_KERNEL);
- if (!c)
- return -ENOMEM;
+ if (c) {
+ spin_lock_init(&c->cnt_lock);
+ spin_lock_init(&c->cs_lock);
+ spin_lock_init(&c->buds_lock);
+ spin_lock_init(&c->space_lock);
+ spin_lock_init(&c->orphan_lock);
+ init_rwsem(&c->commit_sem);
+ mutex_init(&c->lp_mutex);
+ mutex_init(&c->tnc_mutex);
+ mutex_init(&c->log_mutex);
+ mutex_init(&c->mst_mutex);
+ mutex_init(&c->umount_mutex);
+ mutex_init(&c->bu_mutex);
+ mutex_init(&c->write_reserve_mutex);
+ init_waitqueue_head(&c->cmt_wq);
+ c->buds = RB_ROOT;
+ c->old_idx = RB_ROOT;
+ c->size_tree = RB_ROOT;
+ c->orph_tree = RB_ROOT;
+ INIT_LIST_HEAD(&c->infos_list);
+ INIT_LIST_HEAD(&c->idx_gc);
+ INIT_LIST_HEAD(&c->replay_list);
+ INIT_LIST_HEAD(&c->replay_buds);
+ INIT_LIST_HEAD(&c->uncat_list);
+ INIT_LIST_HEAD(&c->empty_list);
+ INIT_LIST_HEAD(&c->freeable_list);
+ INIT_LIST_HEAD(&c->frdi_idx_list);
+ INIT_LIST_HEAD(&c->unclean_leb_list);
+ INIT_LIST_HEAD(&c->old_buds);
+ INIT_LIST_HEAD(&c->orph_list);
+ INIT_LIST_HEAD(&c->orph_new);
+ c->no_chk_data_crc = 1;
+
+ c->highest_inum = UBIFS_FIRST_INO;
+ c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM;
+
+ ubi_get_volume_info(ubi, &c->vi);
+ ubi_get_device_info(c->vi.ubi_num, &c->di);
+ }
+ return c;
+}
- spin_lock_init(&c->cnt_lock);
- spin_lock_init(&c->cs_lock);
- spin_lock_init(&c->buds_lock);
- spin_lock_init(&c->space_lock);
- spin_lock_init(&c->orphan_lock);
- init_rwsem(&c->commit_sem);
- mutex_init(&c->lp_mutex);
- mutex_init(&c->tnc_mutex);
- mutex_init(&c->log_mutex);
- mutex_init(&c->mst_mutex);
- mutex_init(&c->umount_mutex);
- mutex_init(&c->bu_mutex);
- mutex_init(&c->write_reserve_mutex);
- init_waitqueue_head(&c->cmt_wq);
- c->buds = RB_ROOT;
- c->old_idx = RB_ROOT;
- c->size_tree = RB_ROOT;
- c->orph_tree = RB_ROOT;
- INIT_LIST_HEAD(&c->infos_list);
- INIT_LIST_HEAD(&c->idx_gc);
- INIT_LIST_HEAD(&c->replay_list);
- INIT_LIST_HEAD(&c->replay_buds);
- INIT_LIST_HEAD(&c->uncat_list);
- INIT_LIST_HEAD(&c->empty_list);
- INIT_LIST_HEAD(&c->freeable_list);
- INIT_LIST_HEAD(&c->frdi_idx_list);
- INIT_LIST_HEAD(&c->unclean_leb_list);
- INIT_LIST_HEAD(&c->old_buds);
- INIT_LIST_HEAD(&c->orph_list);
- INIT_LIST_HEAD(&c->orph_new);
- c->no_chk_data_crc = 1;
+static int ubifs_fill_super(struct super_block *sb, void *data, int silent)
+{
+ struct ubifs_info *c = sb->s_fs_info;
+ struct inode *root;
+ int err;
c->vfs_sb = sb;
- c->highest_inum = UBIFS_FIRST_INO;
- c->lhead_lnum = c->ltail_lnum = UBIFS_LOG_LNUM;
-
- ubi_get_volume_info(ubi, &c->vi);
- ubi_get_device_info(c->vi.ubi_num, &c->di);
-
/* Re-open the UBI device in read-write mode */
c->ubi = ubi_open_volume(c->vi.ubi_num, c->vi.vol_id, UBI_READWRITE);
if (IS_ERR(c->ubi)) {
err = PTR_ERR(c->ubi);
- goto out_free;
+ goto out;
}
/*
@@ -2091,24 +2094,29 @@ out_bdi:
bdi_destroy(&c->bdi);
out_close:
ubi_close_volume(c->ubi);
-out_free:
- kfree(c);
+out:
return err;
}
static int sb_test(struct super_block *sb, void *data)
{
- dev_t *dev = data;
+ struct ubifs_info *c1 = data;
struct ubifs_info *c = sb->s_fs_info;
- return c->vi.cdev == *dev;
+ return c->vi.cdev == c1->vi.cdev;
+}
+
+static int sb_set(struct super_block *sb, void *data)
+{
+ sb->s_fs_info = data;
+ return set_anon_super(sb, NULL);
}
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;
+ struct ubifs_info *c;
struct super_block *sb;
int err;
@@ -2125,19 +2133,25 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
name, (int)PTR_ERR(ubi));
return ERR_CAST(ubi);
}
- ubi_get_volume_info(ubi, &vi);
- dbg_gen("opened ubi%d_%d", vi.ubi_num, vi.vol_id);
+ c = alloc_ubifs_info(ubi);
+ if (!c) {
+ err = -ENOMEM;
+ goto out_close;
+ }
+
+ dbg_gen("opened ubi%d_%d", c->vi.ubi_num, c->vi.vol_id);
- sb = sget(fs_type, &sb_test, &set_anon_super, &vi.cdev);
+ sb = sget(fs_type, sb_test, sb_set, c);
if (IS_ERR(sb)) {
err = PTR_ERR(sb);
+ kfree(c);
goto out_close;
}
if (sb->s_root) {
struct ubifs_info *c1 = sb->s_fs_info;
-
+ kfree(c);
/* A new mount point for already mounted UBIFS */
dbg_gen("this ubi volume is already mounted");
if (!!(flags & MS_RDONLY) != c1->ro_mount) {
@@ -2146,11 +2160,6 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags,
}
} else {
sb->s_flags = flags;
- /*
- * Pass 'ubi' to 'fill_super()' in sb->s_fs_info where it is
- * replaced by 'c'.
- */
- sb->s_fs_info = ubi;
err = ubifs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0);
if (err)
goto out_deact;
@@ -2170,11 +2179,18 @@ out_close:
return ERR_PTR(err);
}
+static void kill_ubifs_super(struct super_block *s)
+{
+ struct ubifs_info *c = s->s_fs_info;
+ kill_anon_super(s);
+ kfree(c);
+}
+
static struct file_system_type ubifs_fs_type = {
.name = "ubifs",
.owner = THIS_MODULE,
.mount = ubifs_mount,
- .kill_sb = kill_anon_super,
+ .kill_sb = kill_ubifs_super,
};
/*
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index f4213ba1ff85..7f782af286bf 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -131,19 +131,34 @@ xfs_file_fsync(
{
struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
struct xfs_trans *tp;
int error = 0;
int log_flushed = 0;
trace_xfs_file_fsync(ip);
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ if (XFS_FORCED_SHUTDOWN(mp))
return -XFS_ERROR(EIO);
xfs_iflags_clear(ip, XFS_ITRUNCATED);
xfs_ioend_wait(ip);
+ if (mp->m_flags & XFS_MOUNT_BARRIER) {
+ /*
+ * If we have an RT and/or log subvolume we need to make sure
+ * to flush the write cache the device used for file data
+ * first. This is to ensure newly written file data make
+ * it to disk before logging the new inode size in case of
+ * an extending write.
+ */
+ if (XFS_IS_REALTIME_INODE(ip))
+ xfs_blkdev_issue_flush(mp->m_rtdev_targp);
+ else if (mp->m_logdev_targp != mp->m_ddev_targp)
+ xfs_blkdev_issue_flush(mp->m_ddev_targp);
+ }
+
/*
* We always need to make sure that the required inode state is safe on
* disk. The inode might be clean but we still might need to force the
@@ -175,9 +190,9 @@ xfs_file_fsync(
* updates. The sync transaction will also force the log.
*/
xfs_iunlock(ip, XFS_ILOCK_SHARED);
- tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
+ tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
error = xfs_trans_reserve(tp, 0,
- XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
+ XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
if (error) {
xfs_trans_cancel(tp, 0);
return -error;
@@ -209,28 +224,25 @@ xfs_file_fsync(
* force the log.
*/
if (xfs_ipincount(ip)) {
- error = _xfs_log_force_lsn(ip->i_mount,
+ error = _xfs_log_force_lsn(mp,
ip->i_itemp->ili_last_lsn,
XFS_LOG_SYNC, &log_flushed);
}
xfs_iunlock(ip, XFS_ILOCK_SHARED);
}
- if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
- /*
- * If the log write didn't issue an ordered tag we need
- * to flush the disk cache for the data device now.
- */
- if (!log_flushed)
- xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
-
- /*
- * If this inode is on the RT dev we need to flush that
- * cache as well.
- */
- if (XFS_IS_REALTIME_INODE(ip))
- xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
- }
+ /*
+ * If we only have a single device, and the log force about was
+ * a no-op we might have to flush the data device cache here.
+ * This can only happen for fdatasync/O_DSYNC if we were overwriting
+ * an already allocated file and thus do not have any metadata to
+ * commit.
+ */
+ if ((mp->m_flags & XFS_MOUNT_BARRIER) &&
+ mp->m_logdev_targp == mp->m_ddev_targp &&
+ !XFS_IS_REALTIME_INODE(ip) &&
+ !log_flushed)
+ xfs_blkdev_issue_flush(mp->m_ddev_targp);
return -error;
}
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index dd21784525a8..d44d92cd12b1 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -182,7 +182,7 @@ xfs_vn_mknod(
if (IS_POSIXACL(dir)) {
default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT);
if (IS_ERR(default_acl))
- return -PTR_ERR(default_acl);
+ return PTR_ERR(default_acl);
if (!default_acl)
mode &= ~current_umask();
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 1e3a7ce804dc..a1a881e68a9a 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -627,68 +627,6 @@ xfs_blkdev_put(
blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
}
-/*
- * Try to write out the superblock using barriers.
- */
-STATIC int
-xfs_barrier_test(
- xfs_mount_t *mp)
-{
- xfs_buf_t *sbp = xfs_getsb(mp, 0);
- int error;
-
- XFS_BUF_UNDONE(sbp);
- XFS_BUF_UNREAD(sbp);
- XFS_BUF_UNDELAYWRITE(sbp);
- XFS_BUF_WRITE(sbp);
- XFS_BUF_UNASYNC(sbp);
- XFS_BUF_ORDERED(sbp);
-
- xfsbdstrat(mp, sbp);
- error = xfs_buf_iowait(sbp);
-
- /*
- * Clear all the flags we set and possible error state in the
- * buffer. We only did the write to try out whether barriers
- * worked and shouldn't leave any traces in the superblock
- * buffer.
- */
- XFS_BUF_DONE(sbp);
- XFS_BUF_ERROR(sbp, 0);
- XFS_BUF_UNORDERED(sbp);
-
- xfs_buf_relse(sbp);
- return error;
-}
-
-STATIC void
-xfs_mountfs_check_barriers(xfs_mount_t *mp)
-{
- int error;
-
- if (mp->m_logdev_targp != mp->m_ddev_targp) {
- xfs_notice(mp,
- "Disabling barriers, not supported with external log device");
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
- return;
- }
-
- if (xfs_readonly_buftarg(mp->m_ddev_targp)) {
- xfs_notice(mp,
- "Disabling barriers, underlying device is readonly");
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
- return;
- }
-
- error = xfs_barrier_test(mp);
- if (error) {
- xfs_notice(mp,
- "Disabling barriers, trial barrier write failed");
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
- return;
- }
-}
-
void
xfs_blkdev_issue_flush(
xfs_buftarg_t *buftarg)
@@ -1240,14 +1178,6 @@ xfs_fs_remount(
switch (token) {
case Opt_barrier:
mp->m_flags |= XFS_MOUNT_BARRIER;
-
- /*
- * Test if barriers are actually working if we can,
- * else delay this check until the filesystem is
- * marked writeable.
- */
- if (!(mp->m_flags & XFS_MOUNT_RDONLY))
- xfs_mountfs_check_barriers(mp);
break;
case Opt_nobarrier:
mp->m_flags &= ~XFS_MOUNT_BARRIER;
@@ -1282,8 +1212,6 @@ xfs_fs_remount(
/* ro -> rw */
if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) {
mp->m_flags &= ~XFS_MOUNT_RDONLY;
- if (mp->m_flags & XFS_MOUNT_BARRIER)
- xfs_mountfs_check_barriers(mp);
/*
* If this is the first remount to writeable state we
@@ -1465,9 +1393,6 @@ xfs_fs_fill_super(
if (error)
goto out_free_sb;
- if (mp->m_flags & XFS_MOUNT_BARRIER)
- xfs_mountfs_check_barriers(mp);
-
error = xfs_filestream_mount(mp);
if (error)
goto out_free_sb;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 211930246f20..41d5b8f2bf92 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1372,8 +1372,17 @@ xlog_sync(xlog_t *log,
XFS_BUF_ASYNC(bp);
bp->b_flags |= XBF_LOG_BUFFER;
- if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
+ if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {
+ /*
+ * If we have an external log device, flush the data device
+ * before flushing the log to make sure all meta data
+ * written back from the AIL actually made it to disk
+ * before writing out the new log tail LSN in the log buffer.
+ */
+ if (log->l_mp->m_logdev_targp != log->l_mp->m_ddev_targp)
+ xfs_blkdev_issue_flush(log->l_mp->m_ddev_targp);
XFS_BUF_ORDERED(bp);
+ }
ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);