aboutsummaryrefslogtreecommitdiff
path: root/fs/ocfs2
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ocfs2')
-rw-r--r--fs/ocfs2/Kconfig10
-rw-r--r--fs/ocfs2/Makefile7
-rw-r--r--fs/ocfs2/acl.c91
-rw-r--r--fs/ocfs2/acl.h22
-rw-r--r--fs/ocfs2/alloc.c16
-rw-r--r--fs/ocfs2/alloc.h5
-rw-r--r--fs/ocfs2/aops.c38
-rw-r--r--fs/ocfs2/blockcheck.c2
-rw-r--r--fs/ocfs2/buffer_head_io.c2
-rw-r--r--fs/ocfs2/cluster/heartbeat.c12
-rw-r--r--fs/ocfs2/cluster/netdebug.c8
-rw-r--r--fs/ocfs2/cluster/nodemanager.c51
-rw-r--r--fs/ocfs2/cluster/nodemanager.h7
-rw-r--r--fs/ocfs2/cluster/quorum.c16
-rw-r--r--fs/ocfs2/cluster/tcp.c10
-rw-r--r--fs/ocfs2/cluster/tcp_internal.h4
-rw-r--r--fs/ocfs2/dlm/dlmapi.h2
-rw-r--r--fs/ocfs2/dlm/dlmast.c2
-rw-r--r--fs/ocfs2/dlm/dlmconvert.c2
-rw-r--r--fs/ocfs2/dlm/dlmdebug.c2
-rw-r--r--fs/ocfs2/dlm/dlmdomain.c2
-rw-r--r--fs/ocfs2/dlm/dlmlock.c2
-rw-r--r--fs/ocfs2/dlm/dlmmaster.c40
-rw-r--r--fs/ocfs2/dlm/dlmrecovery.c163
-rw-r--r--fs/ocfs2/dlm/dlmunlock.c8
-rw-r--r--fs/ocfs2/dlmglue.c87
-rw-r--r--fs/ocfs2/export.c2
-rw-r--r--fs/ocfs2/extent_map.c27
-rw-r--r--fs/ocfs2/file.c35
-rw-r--r--fs/ocfs2/inode.c4
-rw-r--r--fs/ocfs2/ioctl.c14
-rw-r--r--fs/ocfs2/journal.c4
-rw-r--r--fs/ocfs2/namei.c6
-rw-r--r--fs/ocfs2/ocfs2.h12
-rw-r--r--fs/ocfs2/ocfs2_fs.h13
-rw-r--r--fs/ocfs2/quota.h4
-rw-r--r--fs/ocfs2/quota_local.c2
-rw-r--r--fs/ocfs2/refcounttree.c164
-rw-r--r--fs/ocfs2/stack_o2cb.c12
-rw-r--r--fs/ocfs2/stack_user.c2
-rw-r--r--fs/ocfs2/stackglue.c15
-rw-r--r--fs/ocfs2/super.c97
-rw-r--r--fs/ocfs2/symlink.c12
-rw-r--r--fs/ocfs2/uptodate.c4
-rw-r--r--fs/ocfs2/xattr.c78
-rw-r--r--fs/ocfs2/xattr.h2
46 files changed, 677 insertions, 443 deletions
diff --git a/fs/ocfs2/Kconfig b/fs/ocfs2/Kconfig
index 701b7a3a872..0d840669698 100644
--- a/fs/ocfs2/Kconfig
+++ b/fs/ocfs2/Kconfig
@@ -6,6 +6,7 @@ config OCFS2_FS
select CRC32
select QUOTA
select QUOTA_TREE
+ select FS_POSIX_ACL
help
OCFS2 is a general purpose extent based shared disk cluster file
system with many similarities to ext3. It supports 64 bit inode
@@ -74,12 +75,3 @@ config OCFS2_DEBUG_FS
This option will enable expensive consistency checks. Enable
this option for debugging only as it is likely to decrease
performance of the filesystem.
-
-config OCFS2_FS_POSIX_ACL
- bool "OCFS2 POSIX Access Control Lists"
- depends on OCFS2_FS
- select FS_POSIX_ACL
- default n
- help
- Posix Access Control Lists (ACLs) support permissions for users and
- groups beyond the owner/group/world scheme.
diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile
index 31f25ce32c9..600d2d2ade1 100644
--- a/fs/ocfs2/Makefile
+++ b/fs/ocfs2/Makefile
@@ -39,11 +39,8 @@ ocfs2-objs := \
ver.o \
quota_local.o \
quota_global.o \
- xattr.o
-
-ifeq ($(CONFIG_OCFS2_FS_POSIX_ACL),y)
-ocfs2-objs += acl.o
-endif
+ xattr.o \
+ acl.o
ocfs2_stackglue-objs := stackglue.o
ocfs2_stack_o2cb-objs := stack_o2cb.o
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index fbeaec76210..0501974bedd 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -98,15 +98,11 @@ static struct posix_acl *ocfs2_get_acl_nolock(struct inode *inode,
int type,
struct buffer_head *di_bh)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
int name_index;
char *value = NULL;
struct posix_acl *acl;
int retval;
- if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
- return NULL;
-
switch (type) {
case ACL_TYPE_ACCESS:
name_index = OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS;
@@ -331,13 +327,14 @@ cleanup:
return ret;
}
-static size_t ocfs2_xattr_list_acl_access(struct inode *inode,
+static size_t ocfs2_xattr_list_acl_access(struct dentry *dentry,
char *list,
size_t list_len,
const char *name,
- size_t name_len)
+ size_t name_len,
+ int type)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
@@ -348,13 +345,14 @@ static size_t ocfs2_xattr_list_acl_access(struct inode *inode,
return size;
}
-static size_t ocfs2_xattr_list_acl_default(struct inode *inode,
+static size_t ocfs2_xattr_list_acl_default(struct dentry *dentry,
char *list,
size_t list_len,
const char *name,
- size_t name_len)
+ size_t name_len,
+ int type)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
@@ -365,19 +363,19 @@ static size_t ocfs2_xattr_list_acl_default(struct inode *inode,
return size;
}
-static int ocfs2_xattr_get_acl(struct inode *inode,
- int type,
- void *buffer,
- size_t size)
+static int ocfs2_xattr_get_acl(struct dentry *dentry, const char *name,
+ void *buffer, size_t size, int type)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
struct posix_acl *acl;
int ret;
+ if (strcmp(name, "") != 0)
+ return -EINVAL;
if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
return -EOPNOTSUPP;
- acl = ocfs2_get_acl(inode, type);
+ acl = ocfs2_get_acl(dentry->d_inode, type);
if (IS_ERR(acl))
return PTR_ERR(acl);
if (acl == NULL)
@@ -388,35 +386,16 @@ static int ocfs2_xattr_get_acl(struct inode *inode,
return ret;
}
-static int ocfs2_xattr_get_acl_access(struct inode *inode,
- const char *name,
- void *buffer,
- size_t size)
-{
- if (strcmp(name, "") != 0)
- return -EINVAL;
- return ocfs2_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int ocfs2_xattr_get_acl_default(struct inode *inode,
- const char *name,
- void *buffer,
- size_t size)
-{
- if (strcmp(name, "") != 0)
- return -EINVAL;
- return ocfs2_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int ocfs2_xattr_set_acl(struct inode *inode,
- int type,
- const void *value,
- size_t size)
+static int ocfs2_xattr_set_acl(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags, int type)
{
+ struct inode *inode = dentry->d_inode;
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
struct posix_acl *acl;
int ret = 0;
+ if (strcmp(name, "") != 0)
+ return -EINVAL;
if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
return -EOPNOTSUPP;
@@ -442,38 +421,18 @@ cleanup:
return ret;
}
-static int ocfs2_xattr_set_acl_access(struct inode *inode,
- const char *name,
- const void *value,
- size_t size,
- int flags)
-{
- if (strcmp(name, "") != 0)
- return -EINVAL;
- return ocfs2_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int ocfs2_xattr_set_acl_default(struct inode *inode,
- const char *name,
- const void *value,
- size_t size,
- int flags)
-{
- if (strcmp(name, "") != 0)
- return -EINVAL;
- return ocfs2_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
struct xattr_handler ocfs2_xattr_acl_access_handler = {
.prefix = POSIX_ACL_XATTR_ACCESS,
+ .flags = ACL_TYPE_ACCESS,
.list = ocfs2_xattr_list_acl_access,
- .get = ocfs2_xattr_get_acl_access,
- .set = ocfs2_xattr_set_acl_access,
+ .get = ocfs2_xattr_get_acl,
+ .set = ocfs2_xattr_set_acl,
};
struct xattr_handler ocfs2_xattr_acl_default_handler = {
.prefix = POSIX_ACL_XATTR_DEFAULT,
+ .flags = ACL_TYPE_DEFAULT,
.list = ocfs2_xattr_list_acl_default,
- .get = ocfs2_xattr_get_acl_default,
- .set = ocfs2_xattr_set_acl_default,
+ .get = ocfs2_xattr_get_acl,
+ .set = ocfs2_xattr_set_acl,
};
diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h
index 8f6389ed4da..5c5d31f0585 100644
--- a/fs/ocfs2/acl.h
+++ b/fs/ocfs2/acl.h
@@ -26,8 +26,6 @@ struct ocfs2_acl_entry {
__le32 e_id;
};
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
-
extern int ocfs2_check_acl(struct inode *, int);
extern int ocfs2_acl_chmod(struct inode *);
extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
@@ -35,24 +33,4 @@ extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
struct ocfs2_alloc_context *,
struct ocfs2_alloc_context *);
-#else /* CONFIG_OCFS2_FS_POSIX_ACL*/
-
-#define ocfs2_check_acl NULL
-static inline int ocfs2_acl_chmod(struct inode *inode)
-{
- return 0;
-}
-static inline int ocfs2_init_acl(handle_t *handle,
- struct inode *inode,
- struct inode *dir,
- struct buffer_head *di_bh,
- struct buffer_head *dir_bh,
- struct ocfs2_alloc_context *meta_ac,
- struct ocfs2_alloc_context *data_ac)
-{
- return 0;
-}
-
-#endif /* CONFIG_OCFS2_FS_POSIX_ACL*/
-
#endif /* OCFS2_ACL_H */
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 38a42f5d59f..d17bdc718f7 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -1765,9 +1765,9 @@ set_and_inc:
*
* The array index of the subtree root is passed back.
*/
-static int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
- struct ocfs2_path *left,
- struct ocfs2_path *right)
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+ struct ocfs2_path *left,
+ struct ocfs2_path *right)
{
int i = 0;
@@ -2398,7 +2398,7 @@ static int ocfs2_leftmost_rec_contains(struct ocfs2_extent_list *el, u32 cpos)
*
* The array is assumed to be large enough to hold an entire path (tree depth).
*
- * Upon succesful return from this function:
+ * Upon successful return from this function:
*
* - The 'right_path' array will contain a path to the leaf block
* whose range contains e_cpos.
@@ -2872,8 +2872,8 @@ out:
* This looks similar, but is subtly different to
* ocfs2_find_cpos_for_left_leaf().
*/
-static int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
- struct ocfs2_path *path, u32 *cpos)
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+ struct ocfs2_path *path, u32 *cpos)
{
int i, j, ret = 0;
u64 blkno;
@@ -7190,8 +7190,8 @@ int ocfs2_zero_range_for_truncate(struct inode *inode, handle_t *handle,
* wait on them - the truncate_inode_pages() call later will
* do that for us.
*/
- ret = do_sync_mapping_range(inode->i_mapping, range_start,
- range_end - 1, SYNC_FILE_RANGE_WRITE);
+ ret = filemap_fdatawrite_range(inode->i_mapping, range_start,
+ range_end - 1);
if (ret)
mlog_errno(ret);
diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h
index 9c122d57446..1db4359ccb9 100644
--- a/fs/ocfs2/alloc.h
+++ b/fs/ocfs2/alloc.h
@@ -317,4 +317,9 @@ int ocfs2_path_bh_journal_access(handle_t *handle,
int ocfs2_journal_access_path(struct ocfs2_caching_info *ci,
handle_t *handle,
struct ocfs2_path *path);
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+ struct ocfs2_path *path, u32 *cpos);
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+ struct ocfs2_path *left,
+ struct ocfs2_path *right);
#endif /* OCFS2_ALLOC_H */
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index deb2b132ae5..7e9df11260f 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -547,6 +547,9 @@ bail:
*
* called like this: dio->get_blocks(dio->inode, fs_startblk,
* fs_count, map_bh, dio->rw == WRITE);
+ *
+ * Note that we never bother to allocate blocks here, and thus ignore the
+ * create argument.
*/
static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create)
@@ -563,14 +566,6 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
inode_blocks = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
- /*
- * Any write past EOF is not allowed because we'd be extending.
- */
- if (create && (iblock + max_blocks) > inode_blocks) {
- ret = -EIO;
- goto bail;
- }
-
/* This figures out the size of the next contiguous block, and
* our logical offset */
ret = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno,
@@ -582,15 +577,6 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
goto bail;
}
- if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno && create) {
- ocfs2_error(inode->i_sb,
- "Inode %llu has a hole at block %llu\n",
- (unsigned long long)OCFS2_I(inode)->ip_blkno,
- (unsigned long long)iblock);
- ret = -EROFS;
- goto bail;
- }
-
/* We should already CoW the refcounted extent. */
BUG_ON(ext_flags & OCFS2_EXT_REFCOUNTED);
/*
@@ -601,20 +587,8 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
*/
if (p_blkno && !(ext_flags & OCFS2_EXT_UNWRITTEN))
map_bh(bh_result, inode->i_sb, p_blkno);
- else {
- /*
- * ocfs2_prepare_inode_for_write() should have caught
- * the case where we'd be filling a hole and triggered
- * a buffered write instead.
- */
- if (create) {
- ret = -EIO;
- mlog_errno(ret);
- goto bail;
- }
-
+ else
clear_buffer_mapped(bh_result);
- }
/* make sure we don't map more than max_blocks blocks here as
that's all the kernel will handle at this point. */
@@ -625,7 +599,7 @@ bail:
return ret;
}
-/*
+/*
* ocfs2_dio_end_io is called by the dio core when a dio is finished. We're
* particularly interested in the aio/dio case. Like the core uses
* i_alloc_sem, we use the rw_lock DLM lock to protect io on one node from
@@ -696,7 +670,7 @@ static ssize_t ocfs2_direct_IO(int rw,
ret = blockdev_direct_IO_no_locking(rw, iocb, inode,
inode->i_sb->s_bdev, iov, offset,
- nr_segs,
+ nr_segs,
ocfs2_direct_IO_get_blocks,
ocfs2_dio_end_io);
diff --git a/fs/ocfs2/blockcheck.c b/fs/ocfs2/blockcheck.c
index a1163b8b417..b7428c5d0d3 100644
--- a/fs/ocfs2/blockcheck.c
+++ b/fs/ocfs2/blockcheck.c
@@ -47,7 +47,7 @@
* Calculate the bit offset in the hamming code buffer based on the bit's
* offset in the data buffer. Since the hamming code reserves all
* power-of-two bits for parity, the data bit number and the code bit
- * number are offest by all the parity bits beforehand.
+ * number are offset by all the parity bits beforehand.
*
* Recall that bit numbers in hamming code are 1-based. This function
* takes the 0-based data bit from the caller.
diff --git a/fs/ocfs2/buffer_head_io.c b/fs/ocfs2/buffer_head_io.c
index d43d34a1dd3..21c808f752d 100644
--- a/fs/ocfs2/buffer_head_io.c
+++ b/fs/ocfs2/buffer_head_io.c
@@ -368,7 +368,7 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
}
ocfs2_metadata_cache_io_unlock(ci);
- mlog(ML_BH_IO, "block=(%llu), nr=(%d), cached=%s, flags=0x%x\n",
+ mlog(ML_BH_IO, "block=(%llu), nr=(%d), cached=%s, flags=0x%x\n",
(unsigned long long)block, nr,
((flags & OCFS2_BH_IGNORE_CACHE) || ignore_cache) ? "no" : "yes",
flags);
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index c452d116b89..5c989000670 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -78,7 +78,7 @@ static struct o2hb_callback *hbcall_from_type(enum o2hb_callback_type type);
unsigned int o2hb_dead_threshold = O2HB_DEFAULT_DEAD_THRESHOLD;
-/* Only sets a new threshold if there are no active regions.
+/* Only sets a new threshold if there are no active regions.
*
* No locking or otherwise interesting code is required for reading
* o2hb_dead_threshold as it can't change once regions are active and
@@ -170,13 +170,14 @@ static void o2hb_write_timeout(struct work_struct *work)
mlog(ML_ERROR, "Heartbeat write timeout to device %s after %u "
"milliseconds\n", reg->hr_dev_name,
- jiffies_to_msecs(jiffies - reg->hr_last_timeout_start));
+ jiffies_to_msecs(jiffies - reg->hr_last_timeout_start));
o2quo_disk_timeout();
}
static void o2hb_arm_write_timeout(struct o2hb_region *reg)
{
- mlog(0, "Queue write timeout for %u ms\n", O2HB_MAX_WRITE_TIMEOUT_MS);
+ mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n",
+ O2HB_MAX_WRITE_TIMEOUT_MS);
cancel_delayed_work(&reg->hr_write_timeout_work);
reg->hr_last_timeout_start = jiffies;
@@ -623,7 +624,7 @@ static int o2hb_check_slot(struct o2hb_region *reg,
"seq %llu last %llu changed %u equal %u\n",
slot->ds_node_num, (long long)slot->ds_last_generation,
le32_to_cpu(hb_block->hb_cksum),
- (unsigned long long)le64_to_cpu(hb_block->hb_seq),
+ (unsigned long long)le64_to_cpu(hb_block->hb_seq),
(unsigned long long)slot->ds_last_time, slot->ds_changed_samples,
slot->ds_equal_samples);
@@ -874,7 +875,8 @@ static int o2hb_thread(void *data)
do_gettimeofday(&after_hb);
elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb);
- mlog(0, "start = %lu.%lu, end = %lu.%lu, msec = %u\n",
+ mlog(ML_HEARTBEAT,
+ "start = %lu.%lu, end = %lu.%lu, msec = %u\n",
before_hb.tv_sec, (unsigned long) before_hb.tv_usec,
after_hb.tv_sec, (unsigned long) after_hb.tv_usec,
elapsed_msec);
diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c
index da794bc07a6..a3f150e52b0 100644
--- a/fs/ocfs2/cluster/netdebug.c
+++ b/fs/ocfs2/cluster/netdebug.c
@@ -294,10 +294,10 @@ static int sc_seq_show(struct seq_file *seq, void *v)
if (sc->sc_sock) {
inet = inet_sk(sc->sc_sock->sk);
/* the stack's structs aren't sparse endian clean */
- saddr = (__force __be32)inet->saddr;
- daddr = (__force __be32)inet->daddr;
- sport = (__force __be16)inet->sport;
- dport = (__force __be16)inet->dport;
+ saddr = (__force __be32)inet->inet_saddr;
+ daddr = (__force __be32)inet->inet_daddr;
+ sport = (__force __be16)inet->inet_sport;
+ dport = (__force __be16)inet->inet_dport;
}
/* XXX sigh, inet-> doesn't have sparse annotation so any
diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index 7ee6188bc79..c81142e3ef8 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -35,6 +35,10 @@
* cluster references throughout where nodes are looked up */
struct o2nm_cluster *o2nm_single_cluster = NULL;
+char *o2nm_fence_method_desc[O2NM_FENCE_METHODS] = {
+ "reset", /* O2NM_FENCE_RESET */
+ "panic", /* O2NM_FENCE_PANIC */
+};
struct o2nm_node *o2nm_get_node_by_num(u8 node_num)
{
@@ -579,6 +583,43 @@ static ssize_t o2nm_cluster_attr_reconnect_delay_ms_write(
return o2nm_cluster_attr_write(page, count,
&cluster->cl_reconnect_delay_ms);
}
+
+static ssize_t o2nm_cluster_attr_fence_method_read(
+ struct o2nm_cluster *cluster, char *page)
+{
+ ssize_t ret = 0;
+
+ if (cluster)
+ ret = sprintf(page, "%s\n",
+ o2nm_fence_method_desc[cluster->cl_fence_method]);
+ return ret;
+}
+
+static ssize_t o2nm_cluster_attr_fence_method_write(
+ struct o2nm_cluster *cluster, const char *page, size_t count)
+{
+ unsigned int i;
+
+ if (page[count - 1] != '\n')
+ goto bail;
+
+ for (i = 0; i < O2NM_FENCE_METHODS; ++i) {
+ if (count != strlen(o2nm_fence_method_desc[i]) + 1)
+ continue;
+ if (strncasecmp(page, o2nm_fence_method_desc[i], count - 1))
+ continue;
+ if (cluster->cl_fence_method != i) {
+ printk(KERN_INFO "ocfs2: Changing fence method to %s\n",
+ o2nm_fence_method_desc[i]);
+ cluster->cl_fence_method = i;
+ }
+ return count;
+ }
+
+bail:
+ return -EINVAL;
+}
+
static struct o2nm_cluster_attribute o2nm_cluster_attr_idle_timeout_ms = {
.attr = { .ca_owner = THIS_MODULE,
.ca_name = "idle_timeout_ms",
@@ -603,10 +644,19 @@ static struct o2nm_cluster_attribute o2nm_cluster_attr_reconnect_delay_ms = {
.store = o2nm_cluster_attr_reconnect_delay_ms_write,
};
+static struct o2nm_cluster_attribute o2nm_cluster_attr_fence_method = {
+ .attr = { .ca_owner = THIS_MODULE,
+ .ca_name = "fence_method",
+ .ca_mode = S_IRUGO | S_IWUSR },
+ .show = o2nm_cluster_attr_fence_method_read,
+ .store = o2nm_cluster_attr_fence_method_write,
+};
+
static struct configfs_attribute *o2nm_cluster_attrs[] = {
&o2nm_cluster_attr_idle_timeout_ms.attr,
&o2nm_cluster_attr_keepalive_delay_ms.attr,
&o2nm_cluster_attr_reconnect_delay_ms.attr,
+ &o2nm_cluster_attr_fence_method.attr,
NULL,
};
static ssize_t o2nm_cluster_show(struct config_item *item,
@@ -778,6 +828,7 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
cluster->cl_reconnect_delay_ms = O2NET_RECONNECT_DELAY_MS_DEFAULT;
cluster->cl_idle_timeout_ms = O2NET_IDLE_TIMEOUT_MS_DEFAULT;
cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT;
+ cluster->cl_fence_method = O2NM_FENCE_RESET;
ret = &cluster->cl_group;
o2nm_single_cluster = cluster;
diff --git a/fs/ocfs2/cluster/nodemanager.h b/fs/ocfs2/cluster/nodemanager.h
index c992ea0da4a..09ea2d388bb 100644
--- a/fs/ocfs2/cluster/nodemanager.h
+++ b/fs/ocfs2/cluster/nodemanager.h
@@ -33,6 +33,12 @@
#include <linux/configfs.h>
#include <linux/rbtree.h>
+enum o2nm_fence_method {
+ O2NM_FENCE_RESET = 0,
+ O2NM_FENCE_PANIC,
+ O2NM_FENCE_METHODS, /* Number of fence methods */
+};
+
struct o2nm_node {
spinlock_t nd_lock;
struct config_item nd_item;
@@ -58,6 +64,7 @@ struct o2nm_cluster {
unsigned int cl_idle_timeout_ms;
unsigned int cl_keepalive_delay_ms;
unsigned int cl_reconnect_delay_ms;
+ enum o2nm_fence_method cl_fence_method;
/* this bitmap is part of a hack for disk bitmap.. will go eventually. - zab */
unsigned long cl_nodes_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c
index bbacf7da48a..639024033fc 100644
--- a/fs/ocfs2/cluster/quorum.c
+++ b/fs/ocfs2/cluster/quorum.c
@@ -74,8 +74,20 @@ static void o2quo_fence_self(void)
* threads can still schedule, etc, etc */
o2hb_stop_all_regions();
- printk("ocfs2 is very sorry to be fencing this system by restarting\n");
- emergency_restart();
+ switch (o2nm_single_cluster->cl_fence_method) {
+ case O2NM_FENCE_PANIC:
+ panic("*** ocfs2 is very sorry to be fencing this system by "
+ "panicing ***\n");
+ break;
+ default:
+ WARN_ON(o2nm_single_cluster->cl_fence_method >=
+ O2NM_FENCE_METHODS);
+ case O2NM_FENCE_RESET:
+ printk(KERN_ERR "*** ocfs2 is very sorry to be fencing this "
+ "system by restarting ***\n");
+ emergency_restart();
+ break;
+ };
}
/* Indicate that a timeout occured on a hearbeat region write. The
diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c
index 334f231a422..d8d0c65ac03 100644
--- a/fs/ocfs2/cluster/tcp.c
+++ b/fs/ocfs2/cluster/tcp.c
@@ -485,7 +485,7 @@ static void o2net_set_nn_state(struct o2net_node *nn,
}
if (was_valid && !valid) {
- printk(KERN_INFO "o2net: no longer connected to "
+ printk(KERN_NOTICE "o2net: no longer connected to "
SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc));
o2net_complete_nodes_nsw(nn);
}
@@ -493,7 +493,7 @@ static void o2net_set_nn_state(struct o2net_node *nn,
if (!was_valid && valid) {
o2quo_conn_up(o2net_num_from_nn(nn));
cancel_delayed_work(&nn->nn_connect_expired);
- printk(KERN_INFO "o2net: %s " SC_NODEF_FMT "\n",
+ printk(KERN_NOTICE "o2net: %s " SC_NODEF_FMT "\n",
o2nm_this_node() > sc->sc_node->nd_num ?
"connected to" : "accepted connection from",
SC_NODEF_ARGS(sc));
@@ -930,7 +930,7 @@ static void o2net_sendpage(struct o2net_sock_container *sc,
cond_resched();
continue;
}
- mlog(ML_ERROR, "sendpage of size %zu to " SC_NODEF_FMT
+ mlog(ML_ERROR, "sendpage of size %zu to " SC_NODEF_FMT
" failed with %zd\n", size, SC_NODEF_ARGS(sc), ret);
o2net_ensure_shutdown(nn, sc, 0);
break;
@@ -1476,14 +1476,14 @@ static void o2net_idle_timer(unsigned long data)
do_gettimeofday(&now);
- printk(KERN_INFO "o2net: connection to " SC_NODEF_FMT " has been idle for %u.%u "
+ printk(KERN_NOTICE "o2net: connection to " SC_NODEF_FMT " has been idle for %u.%u "
"seconds, shutting it down.\n", SC_NODEF_ARGS(sc),
o2net_idle_timeout() / 1000,
o2net_idle_timeout() % 1000);
mlog(ML_NOTICE, "here are some times that might help debug the "
"situation: (tmr %ld.%ld now %ld.%ld dr %ld.%ld adv "
"%ld.%ld:%ld.%ld func (%08x:%u) %ld.%ld:%ld.%ld)\n",
- sc->sc_tv_timer.tv_sec, (long) sc->sc_tv_timer.tv_usec,
+ sc->sc_tv_timer.tv_sec, (long) sc->sc_tv_timer.tv_usec,
now.tv_sec, (long) now.tv_usec,
sc->sc_tv_data_ready.tv_sec, (long) sc->sc_tv_data_ready.tv_usec,
sc->sc_tv_advance_start.tv_sec,
diff --git a/fs/ocfs2/cluster/tcp_internal.h b/fs/ocfs2/cluster/tcp_internal.h
index 8d58cfe410b..96fa7ebc530 100644
--- a/fs/ocfs2/cluster/tcp_internal.h
+++ b/fs/ocfs2/cluster/tcp_internal.h
@@ -32,10 +32,10 @@
* on their number */
#define O2NET_QUORUM_DELAY_MS ((o2hb_dead_threshold + 2) * O2HB_REGION_TIMEOUT_MS)
-/*
+/*
* This version number represents quite a lot, unfortunately. It not
* only represents the raw network message protocol on the wire but also
- * locking semantics of the file system using the protocol. It should
+ * locking semantics of the file system using the protocol. It should
* be somewhere else, I'm sure, but right now it isn't.
*
* With version 11, we separate out the filesystem locking portion. The
diff --git a/fs/ocfs2/dlm/dlmapi.h b/fs/ocfs2/dlm/dlmapi.h
index b5786a787fa..3cfa114aa39 100644
--- a/fs/ocfs2/dlm/dlmapi.h
+++ b/fs/ocfs2/dlm/dlmapi.h
@@ -95,7 +95,7 @@ const char *dlm_errname(enum dlm_status err);
mlog(ML_ERROR, "dlm status = %s\n", dlm_errname((st))); \
} while (0)
-#define DLM_LKSB_UNUSED1 0x01
+#define DLM_LKSB_UNUSED1 0x01
#define DLM_LKSB_PUT_LVB 0x02
#define DLM_LKSB_GET_LVB 0x04
#define DLM_LKSB_UNUSED2 0x08
diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c
index 01cf8cc3d28..dccc439fa08 100644
--- a/fs/ocfs2/dlm/dlmast.c
+++ b/fs/ocfs2/dlm/dlmast.c
@@ -123,7 +123,7 @@ static void __dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock)
dlm_lock_put(lock);
/* free up the reserved bast that we are cancelling.
* guaranteed that this will not be the last reserved
- * ast because *both* an ast and a bast were reserved
+ * ast because *both* an ast and a bast were reserved
* to get to this point. the res->spinlock will not be
* taken here */
dlm_lockres_release_ast(dlm, res);
diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c
index ca96bce50e1..f283bce776b 100644
--- a/fs/ocfs2/dlm/dlmconvert.c
+++ b/fs/ocfs2/dlm/dlmconvert.c
@@ -396,7 +396,7 @@ static enum dlm_status dlm_send_remote_convert_request(struct dlm_ctxt *dlm,
/* instead of logging the same network error over
* and over, sleep here and wait for the heartbeat
* to notice the node is dead. times out after 5s. */
- dlm_wait_for_node_death(dlm, res->owner,
+ dlm_wait_for_node_death(dlm, res->owner,
DLM_NODE_DEATH_WAIT_MAX);
ret = DLM_RECOVERING;
mlog(0, "node %u died so returning DLM_RECOVERING "
diff --git a/fs/ocfs2/dlm/dlmdebug.c b/fs/ocfs2/dlm/dlmdebug.c
index 42b0bad7a61..0cd24cf5439 100644
--- a/fs/ocfs2/dlm/dlmdebug.c
+++ b/fs/ocfs2/dlm/dlmdebug.c
@@ -102,7 +102,7 @@ void __dlm_print_one_lock_resource(struct dlm_lock_resource *res)
assert_spin_locked(&res->spinlock);
stringify_lockname(res->lockname.name, res->lockname.len,
- buf, sizeof(buf) - 1);
+ buf, sizeof(buf));
printk("lockres: %s, owner=%u, state=%u\n",
buf, res->owner, res->state);
printk(" last used: %lu, refcnt: %u, on purge list: %s\n",
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 0334000676d..988c9055fd4 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -816,7 +816,7 @@ static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data,
}
/* Once the dlm ctxt is marked as leaving then we don't want
- * to be put in someone's domain map.
+ * to be put in someone's domain map.
* Also, explicitly disallow joining at certain troublesome
* times (ie. during recovery). */
if (dlm && dlm->dlm_state != DLM_CTXT_LEAVING) {
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index 437698e9465..73333777267 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -269,7 +269,7 @@ static enum dlm_status dlmlock_remote(struct dlm_ctxt *dlm,
}
dlm_revert_pending_lock(res, lock);
dlm_lock_put(lock);
- } else if (dlm_is_recovery_lock(res->lockname.name,
+ } else if (dlm_is_recovery_lock(res->lockname.name,
res->lockname.len)) {
/* special case for the $RECOVERY lock.
* there will never be an AST delivered to put
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 83bcaf266b3..a659606dcb9 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -366,7 +366,7 @@ void dlm_hb_event_notify_attached(struct dlm_ctxt *dlm, int idx, int node_up)
struct dlm_master_list_entry *mle;
assert_spin_locked(&dlm->spinlock);
-
+
list_for_each_entry(mle, &dlm->mle_hb_events, hb_events) {
if (node_up)
dlm_mle_node_up(dlm, mle, NULL, idx);
@@ -833,7 +833,7 @@ lookup:
__dlm_insert_mle(dlm, mle);
/* still holding the dlm spinlock, check the recovery map
- * to see if there are any nodes that still need to be
+ * to see if there are any nodes that still need to be
* considered. these will not appear in the mle nodemap
* but they might own this lockres. wait on them. */
bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0);
@@ -883,7 +883,7 @@ redo_request:
msleep(500);
}
continue;
- }
+ }
dlm_kick_recovery_thread(dlm);
msleep(1000);
@@ -939,8 +939,8 @@ wait:
res->lockname.name, blocked);
if (++tries > 20) {
mlog(ML_ERROR, "%s:%.*s: spinning on "
- "dlm_wait_for_lock_mastery, blocked=%d\n",
- dlm->name, res->lockname.len,
+ "dlm_wait_for_lock_mastery, blocked=%d\n",
+ dlm->name, res->lockname.len,
res->lockname.name, blocked);
dlm_print_one_lock_resource(res);
dlm_print_one_mle(mle);
@@ -1029,7 +1029,7 @@ recheck:
ret = dlm_restart_lock_mastery(dlm, res, mle, *blocked);
b = (mle->type == DLM_MLE_BLOCK);
if ((*blocked && !b) || (!*blocked && b)) {
- mlog(0, "%s:%.*s: status change: old=%d new=%d\n",
+ mlog(0, "%s:%.*s: status change: old=%d new=%d\n",
dlm->name, res->lockname.len, res->lockname.name,
*blocked, b);
*blocked = b;
@@ -1602,7 +1602,7 @@ send_response:
}
mlog(0, "%u is the owner of %.*s, cleaning everyone else\n",
dlm->node_num, res->lockname.len, res->lockname.name);
- ret = dlm_dispatch_assert_master(dlm, res, 0, request->node_idx,
+ ret = dlm_dispatch_assert_master(dlm, res, 0, request->node_idx,
DLM_ASSERT_MASTER_MLE_CLEANUP);
if (ret < 0) {
mlog(ML_ERROR, "failed to dispatch assert master work\n");
@@ -1701,7 +1701,7 @@ again:
if (r & DLM_ASSERT_RESPONSE_REASSERT) {
mlog(0, "%.*s: node %u create mles on other "
- "nodes and requests a re-assert\n",
+ "nodes and requests a re-assert\n",
namelen, lockname, to);
reassert = 1;
}
@@ -1812,7 +1812,7 @@ int dlm_assert_master_handler(struct o2net_msg *msg, u32 len, void *data,
spin_unlock(&dlm->master_lock);
spin_unlock(&dlm->spinlock);
goto done;
- }
+ }
}
}
spin_unlock(&dlm->master_lock);
@@ -1883,7 +1883,7 @@ ok:
int extra_ref = 0;
int nn = -1;
int rr, err = 0;
-
+
spin_lock(&mle->spinlock);
if (mle->type == DLM_MLE_BLOCK || mle->type == DLM_MLE_MIGRATION)
extra_ref = 1;
@@ -1891,7 +1891,7 @@ ok:
/* MASTER mle: if any bits set in the response map
* then the calling node needs to re-assert to clear
* up nodes that this node contacted */
- while ((nn = find_next_bit (mle->response_map, O2NM_MAX_NODES,
+ while ((nn = find_next_bit (mle->response_map, O2NM_MAX_NODES,
nn+1)) < O2NM_MAX_NODES) {
if (nn != dlm->node_num && nn != assert->node_idx)
master_request = 1;
@@ -2002,7 +2002,7 @@ kill:
__dlm_print_one_lock_resource(res);
spin_unlock(&res->spinlock);
spin_unlock(&dlm->spinlock);
- *ret_data = (void *)res;
+ *ret_data = (void *)res;
dlm_put(dlm);
return -EINVAL;
}
@@ -2040,10 +2040,10 @@ int dlm_dispatch_assert_master(struct dlm_ctxt *dlm,
item->u.am.request_from = request_from;
item->u.am.flags = flags;
- if (ignore_higher)
- mlog(0, "IGNORE HIGHER: %.*s\n", res->lockname.len,
+ if (ignore_higher)
+ mlog(0, "IGNORE HIGHER: %.*s\n", res->lockname.len,
res->lockname.name);
-
+
spin_lock(&dlm->work_lock);
list_add_tail(&item->list, &dlm->work_list);
spin_unlock(&dlm->work_lock);
@@ -2133,7 +2133,7 @@ put:
* think that $RECOVERY is currently mastered by a dead node. If so,
* we wait a short time to allow that node to get notified by its own
* heartbeat stack, then check again. All $RECOVERY lock resources
- * mastered by dead nodes are purged when the hearbeat callback is
+ * mastered by dead nodes are purged when the hearbeat callback is
* fired, so we can know for sure that it is safe to continue once
* the node returns a live node or no node. */
static int dlm_pre_master_reco_lockres(struct dlm_ctxt *dlm,
@@ -2174,7 +2174,7 @@ static int dlm_pre_master_reco_lockres(struct dlm_ctxt *dlm,
ret = -EAGAIN;
}
spin_unlock(&dlm->spinlock);
- mlog(0, "%s: reco lock master is %u\n", dlm->name,
+ mlog(0, "%s: reco lock master is %u\n", dlm->name,
master);
break;
}
@@ -2586,7 +2586,7 @@ fail:
* is complete everywhere. if the target dies while this is
* going on, some nodes could potentially see the target as the
* master, so it is important that my recovery finds the migration
- * mle and sets the master to UNKNONWN. */
+ * mle and sets the master to UNKNOWN. */
/* wait for new node to assert master */
@@ -2602,7 +2602,7 @@ fail:
mlog(0, "%s:%.*s: timed out during migration\n",
dlm->name, res->lockname.len, res->lockname.name);
- /* avoid hang during shutdown when migrating lockres
+ /* avoid hang during shutdown when migrating lockres
* to a node which also goes down */
if (dlm_is_node_dead(dlm, target)) {
mlog(0, "%s:%.*s: expected migration "
@@ -2738,7 +2738,7 @@ static int dlm_migration_can_proceed(struct dlm_ctxt *dlm,
can_proceed = !!(res->state & DLM_LOCK_RES_MIGRATING);
spin_unlock(&res->spinlock);
- /* target has died, so make the caller break out of the
+ /* target has died, so make the caller break out of the
* wait_event, but caller must recheck the domain_map */
spin_lock(&dlm->spinlock);
if (!test_bit(mig_target, dlm->domain_map))
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index d9fa3d22e17..344bcf90cbf 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -1050,7 +1050,7 @@ static void dlm_move_reco_locks_to_list(struct dlm_ctxt *dlm,
if (lock->ml.node == dead_node) {
mlog(0, "AHA! there was "
"a $RECOVERY lock for dead "
- "node %u (%s)!\n",
+ "node %u (%s)!\n",
dead_node, dlm->name);
list_del_init(&lock->list);
dlm_lock_put(lock);
@@ -1164,6 +1164,39 @@ static void dlm_init_migratable_lockres(struct dlm_migratable_lockres *mres,
mres->master = master;
}
+static void dlm_prepare_lvb_for_migration(struct dlm_lock *lock,
+ struct dlm_migratable_lockres *mres,
+ int queue)
+{
+ if (!lock->lksb)
+ return;
+
+ /* Ignore lvb in all locks in the blocked list */
+ if (queue == DLM_BLOCKED_LIST)
+ return;
+
+ /* Only consider lvbs in locks with granted EX or PR lock levels */
+ if (lock->ml.type != LKM_EXMODE && lock->ml.type != LKM_PRMODE)
+ return;
+
+ if (dlm_lvb_is_empty(mres->lvb)) {
+ memcpy(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN);
+ return;
+ }
+
+ /* Ensure the lvb copied for migration matches in other valid locks */
+ if (!memcmp(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN))
+ return;
+
+ mlog(ML_ERROR, "Mismatched lvb in lock cookie=%u:%llu, name=%.*s, "
+ "node=%u\n",
+ dlm_get_lock_cookie_node(be64_to_cpu(lock->ml.cookie)),
+ dlm_get_lock_cookie_seq(be64_to_cpu(lock->ml.cookie)),
+ lock->lockres->lockname.len, lock->lockres->lockname.name,
+ lock->ml.node);
+ dlm_print_one_lock_resource(lock->lockres);
+ BUG();
+}
/* returns 1 if this lock fills the network structure,
* 0 otherwise */
@@ -1181,20 +1214,7 @@ static int dlm_add_lock_to_array(struct dlm_lock *lock,
ml->list = queue;
if (lock->lksb) {
ml->flags = lock->lksb->flags;
- /* send our current lvb */
- if (ml->type == LKM_EXMODE ||
- ml->type == LKM_PRMODE) {
- /* if it is already set, this had better be a PR
- * and it has to match */
- if (!dlm_lvb_is_empty(mres->lvb) &&
- (ml->type == LKM_EXMODE ||
- memcmp(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN))) {
- mlog(ML_ERROR, "mismatched lvbs!\n");
- dlm_print_one_lock_resource(lock->lockres);
- BUG();
- }
- memcpy(mres->lvb, lock->lksb->lvb, DLM_LVB_LEN);
- }
+ dlm_prepare_lvb_for_migration(lock, mres, queue);
}
ml->node = lock->ml.node;
mres->num_locks++;
@@ -1730,6 +1750,7 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm,
struct dlm_lock *lock = NULL;
u8 from = O2NM_MAX_NODES;
unsigned int added = 0;
+ __be64 c;
mlog(0, "running %d locks for this lockres\n", mres->num_locks);
for (i=0; i<mres->num_locks; i++) {
@@ -1777,19 +1798,48 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm,
/* lock is always created locally first, and
* destroyed locally last. it must be on the list */
if (!lock) {
- __be64 c = ml->cookie;
- mlog(ML_ERROR, "could not find local lock "
- "with cookie %u:%llu!\n",
+ c = ml->cookie;
+ mlog(ML_ERROR, "Could not find local lock "
+ "with cookie %u:%llu, node %u, "
+ "list %u, flags 0x%x, type %d, "
+ "conv %d, highest blocked %d\n",
dlm_get_lock_cookie_node(be64_to_cpu(c)),
- dlm_get_lock_cookie_seq(be64_to_cpu(c)));
+ dlm_get_lock_cookie_seq(be64_to_cpu(c)),
+ ml->node, ml->list, ml->flags, ml->type,
+ ml->convert_type, ml->highest_blocked);
+ __dlm_print_one_lock_resource(res);
+ BUG();
+ }
+
+ if (lock->ml.node != ml->node) {
+ c = lock->ml.cookie;
+ mlog(ML_ERROR, "Mismatched node# in lock "
+ "cookie %u:%llu, name %.*s, node %u\n",
+ dlm_get_lock_cookie_node(be64_to_cpu(c)),
+ dlm_get_lock_cookie_seq(be64_to_cpu(c)),
+ res->lockname.len, res->lockname.name,
+ lock->ml.node);
+ c = ml->cookie;
+ mlog(ML_ERROR, "Migrate lock cookie %u:%llu, "
+ "node %u, list %u, flags 0x%x, type %d, "
+ "conv %d, highest blocked %d\n",
+ dlm_get_lock_cookie_node(be64_to_cpu(c)),
+ dlm_get_lock_cookie_seq(be64_to_cpu(c)),
+ ml->node, ml->list, ml->flags, ml->type,
+ ml->convert_type, ml->highest_blocked);
__dlm_print_one_lock_resource(res);
BUG();
}
- BUG_ON(lock->ml.node != ml->node);
if (tmpq != queue) {
- mlog(0, "lock was on %u instead of %u for %.*s\n",
- j, ml->list, res->lockname.len, res->lockname.name);
+ c = ml->cookie;
+ mlog(0, "Lock cookie %u:%llu was on list %u "
+ "instead of list %u for %.*s\n",
+ dlm_get_lock_cookie_node(be64_to_cpu(c)),
+ dlm_get_lock_cookie_seq(be64_to_cpu(c)),
+ j, ml->list, res->lockname.len,
+ res->lockname.name);
+ __dlm_print_one_lock_resource(res);
spin_unlock(&res->spinlock);
continue;
}
@@ -1839,7 +1889,7 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm,
* the lvb. */
memcpy(res->lvb, mres->lvb, DLM_LVB_LEN);
} else {
- /* otherwise, the node is sending its
+ /* otherwise, the node is sending its
* most recent valid lvb info */
BUG_ON(ml->type != LKM_EXMODE &&
ml->type != LKM_PRMODE);
@@ -1886,7 +1936,7 @@ skip_lvb:
spin_lock(&res->spinlock);
list_for_each_entry(lock, queue, list) {
if (lock->ml.cookie == ml->cookie) {
- __be64 c = lock->ml.cookie;
+ c = lock->ml.cookie;
mlog(ML_ERROR, "%s:%.*s: %u:%llu: lock already "
"exists on this lockres!\n", dlm->name,
res->lockname.len, res->lockname.name,
@@ -2114,7 +2164,7 @@ static void dlm_revalidate_lvb(struct dlm_ctxt *dlm,
assert_spin_locked(&res->spinlock);
if (res->owner == dlm->node_num)
- /* if this node owned the lockres, and if the dead node
+ /* if this node owned the lockres, and if the dead node
* had an EX when he died, blank out the lvb */
search_node = dead_node;
else {
@@ -2152,7 +2202,7 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm,
/* this node is the lockres master:
* 1) remove any stale locks for the dead node
- * 2) if the dead node had an EX when he died, blank out the lvb
+ * 2) if the dead node had an EX when he died, blank out the lvb
*/
assert_spin_locked(&dlm->spinlock);
assert_spin_locked(&res->spinlock);
@@ -2193,7 +2243,12 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm,
mlog(0, "%s:%.*s: freed %u locks for dead node %u, "
"dropping ref from lockres\n", dlm->name,
res->lockname.len, res->lockname.name, freed, dead_node);
- BUG_ON(!test_bit(dead_node, res->refmap));
+ if(!test_bit(dead_node, res->refmap)) {
+ mlog(ML_ERROR, "%s:%.*s: freed %u locks for dead node %u, "
+ "but ref was not set\n", dlm->name,
+ res->lockname.len, res->lockname.name, freed, dead_node);
+ __dlm_print_one_lock_resource(res);
+ }
dlm_lockres_clear_refmap_bit(dead_node, res);
} else if (test_bit(dead_node, res->refmap)) {
mlog(0, "%s:%.*s: dead node %u had a ref, but had "
@@ -2260,7 +2315,7 @@ static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node)
}
spin_unlock(&res->spinlock);
continue;
- }
+ }
spin_lock(&res->spinlock);
/* zero the lvb if necessary */
dlm_revalidate_lvb(dlm, res, dead_node);
@@ -2411,7 +2466,7 @@ static void dlm_reco_unlock_ast(void *astdata, enum dlm_status st)
* this function on each node racing to become the recovery
* master will not stop attempting this until either:
* a) this node gets the EX (and becomes the recovery master),
- * or b) dlm->reco.new_master gets set to some nodenum
+ * or b) dlm->reco.new_master gets set to some nodenum
* != O2NM_INVALID_NODE_NUM (another node will do the reco).
* so each time a recovery master is needed, the entire cluster
* will sync at this point. if the new master dies, that will
@@ -2424,7 +2479,7 @@ static int dlm_pick_recovery_master(struct dlm_ctxt *dlm)
mlog(0, "starting recovery of %s at %lu, dead=%u, this=%u\n",
dlm->name, jiffies, dlm->reco.dead_node, dlm->node_num);
-again:
+again:
memset(&lksb, 0, sizeof(lksb));
ret = dlmlock(dlm, LKM_EXMODE, &lksb, LKM_NOQUEUE|LKM_RECOVERY,
@@ -2437,8 +2492,8 @@ again:
if (ret == DLM_NORMAL) {
mlog(0, "dlm=%s dlmlock says I got it (this=%u)\n",
dlm->name, dlm->node_num);
-
- /* got the EX lock. check to see if another node
+
+ /* got the EX lock. check to see if another node
* just became the reco master */
if (dlm_reco_master_ready(dlm)) {
mlog(0, "%s: got reco EX lock, but %u will "
@@ -2451,12 +2506,12 @@ again:
/* see if recovery was already finished elsewhere */
spin_lock(&dlm->spinlock);
if (dlm->reco.dead_node == O2NM_INVALID_NODE_NUM) {
- status = -EINVAL;
+ status = -EINVAL;
mlog(0, "%s: got reco EX lock, but "
"node got recovered already\n", dlm->name);
if (dlm->reco.new_master != O2NM_INVALID_NODE_NUM) {
mlog(ML_ERROR, "%s: new master is %u "
- "but no dead node!\n",
+ "but no dead node!\n",
dlm->name, dlm->reco.new_master);
BUG();
}
@@ -2468,7 +2523,7 @@ again:
* set the master and send the messages to begin recovery */
if (!status) {
mlog(0, "%s: dead=%u, this=%u, sending "
- "begin_reco now\n", dlm->name,
+ "begin_reco now\n", dlm->name,
dlm->reco.dead_node, dlm->node_num);
status = dlm_send_begin_reco_message(dlm,
dlm->reco.dead_node);
@@ -2501,7 +2556,7 @@ again:
mlog(0, "dlm=%s dlmlock says another node got it (this=%u)\n",
dlm->name, dlm->node_num);
/* another node is master. wait on
- * reco.new_master != O2NM_INVALID_NODE_NUM
+ * reco.new_master != O2NM_INVALID_NODE_NUM
* for at most one second */
wait_event_timeout(dlm->dlm_reco_thread_wq,
dlm_reco_master_ready(dlm),
@@ -2589,9 +2644,23 @@ retry:
"begin reco msg (%d)\n", dlm->name, nodenum, ret);
ret = 0;
}
+
+ /*
+ * Prior to commit aad1b15310b9bcd59fa81ab8f2b1513b59553ea8,
+ * dlm_begin_reco_handler() returned EAGAIN and not -EAGAIN.
+ * We are handling both for compatibility reasons.
+ */
+ if (ret == -EAGAIN || ret == EAGAIN) {
+ mlog(0, "%s: trying to start recovery of node "
+ "%u, but node %u is waiting for last recovery "
+ "to complete, backoff for a bit\n", dlm->name,
+ dead_node, nodenum);
+ msleep(100);
+ goto retry;
+ }
if (ret < 0) {
struct dlm_lock_resource *res;
- /* this is now a serious problem, possibly ENOMEM
+ /* this is now a serious problem, possibly ENOMEM
* in the network stack. must retry */
mlog_errno(ret);
mlog(ML_ERROR, "begin reco of dlm %s to node %u "
@@ -2604,18 +2673,10 @@ retry:
} else {
mlog(ML_ERROR, "recovery lock not found\n");
}
- /* sleep for a bit in hopes that we can avoid
+ /* sleep for a bit in hopes that we can avoid
* another ENOMEM */
msleep(100);
goto retry;
- } else if (ret == EAGAIN) {
- mlog(0, "%s: trying to start recovery of node "
- "%u, but node %u is waiting for last recovery "
- "to complete, backoff for a bit\n", dlm->name,
- dead_node, nodenum);
- /* TODO Look into replacing msleep with cond_resched() */
- msleep(100);
- goto retry;
}
}
@@ -2639,7 +2700,7 @@ int dlm_begin_reco_handler(struct o2net_msg *msg, u32 len, void *data,
dlm->name, br->node_idx, br->dead_node,
dlm->reco.dead_node, dlm->reco.new_master);
spin_unlock(&dlm->spinlock);
- return EAGAIN;
+ return -EAGAIN;
}
spin_unlock(&dlm->spinlock);
@@ -2664,7 +2725,7 @@ int dlm_begin_reco_handler(struct o2net_msg *msg, u32 len, void *data,
}
if (dlm->reco.dead_node != O2NM_INVALID_NODE_NUM) {
mlog(ML_NOTICE, "%s: dead_node previously set to %u, "
- "node %u changing it to %u\n", dlm->name,
+ "node %u changing it to %u\n", dlm->name,
dlm->reco.dead_node, br->node_idx, br->dead_node);
}
dlm_set_reco_master(dlm, br->node_idx);
@@ -2730,8 +2791,8 @@ stage2:
if (ret < 0) {
mlog_errno(ret);
if (dlm_is_host_down(ret)) {
- /* this has no effect on this recovery
- * session, so set the status to zero to
+ /* this has no effect on this recovery
+ * session, so set the status to zero to
* finish out the last recovery */
mlog(ML_ERROR, "node %u went down after this "
"node finished recovery.\n", nodenum);
@@ -2768,7 +2829,7 @@ int dlm_finalize_reco_handler(struct o2net_msg *msg, u32 len, void *data,
mlog(0, "%s: node %u finalizing recovery stage%d of "
"node %u (%u:%u)\n", dlm->name, fr->node_idx, stage,
fr->dead_node, dlm->reco.dead_node, dlm->reco.new_master);
-
+
spin_lock(&dlm->spinlock);
if (dlm->reco.new_master != fr->node_idx) {
diff --git a/fs/ocfs2/dlm/dlmunlock.c b/fs/ocfs2/dlm/dlmunlock.c
index 00f53b2aea7..49e29ecd020 100644
--- a/fs/ocfs2/dlm/dlmunlock.c
+++ b/fs/ocfs2/dlm/dlmunlock.c
@@ -190,8 +190,8 @@ static enum dlm_status dlmunlock_common(struct dlm_ctxt *dlm,
actions &= ~(DLM_UNLOCK_REMOVE_LOCK|
DLM_UNLOCK_REGRANT_LOCK|
DLM_UNLOCK_CLEAR_CONVERT_TYPE);
- } else if (status == DLM_RECOVERING ||
- status == DLM_MIGRATING ||
+ } else if (status == DLM_RECOVERING ||
+ status == DLM_MIGRATING ||
status == DLM_FORWARD) {
/* must clear the actions because this unlock
* is about to be retried. cannot free or do
@@ -661,14 +661,14 @@ retry:
if (call_ast) {
mlog(0, "calling unlockast(%p, %d)\n", data, status);
if (is_master) {
- /* it is possible that there is one last bast
+ /* it is possible that there is one last bast
* pending. make sure it is flushed, then
* call the unlockast.
* not an issue if this is a mastered remotely,
* since this lock has been removed from the
* lockres queues and cannot be found. */
dlm_kick_thread(dlm, NULL);
- wait_event(dlm->ast_wq,
+ wait_event(dlm->ast_wq,
dlm_lock_basts_flushed(dlm, lock));
}
(*unlockast)(data, status);
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 0d38d67194c..e044019cb3b 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -875,6 +875,14 @@ static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lo
lockres_or_flags(lockres, OCFS2_LOCK_NEEDS_REFRESH);
lockres->l_level = lockres->l_requested;
+
+ /*
+ * We set the OCFS2_LOCK_UPCONVERT_FINISHING flag before clearing
+ * the OCFS2_LOCK_BUSY flag to prevent the dc thread from
+ * downconverting the lock before the upconvert has fully completed.
+ */
+ lockres_or_flags(lockres, OCFS2_LOCK_UPCONVERT_FINISHING);
+
lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
mlog_exit_void();
@@ -907,8 +915,6 @@ static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
assert_spin_locked(&lockres->l_lock);
- lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED);
-
if (level > lockres->l_blocking) {
/* only schedule a downconvert if we haven't already scheduled
* one that goes low enough to satisfy the level we're
@@ -921,6 +927,9 @@ static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
lockres->l_blocking = level;
}
+ if (needs_downconvert)
+ lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED);
+
mlog_exit(needs_downconvert);
return needs_downconvert;
}
@@ -1133,6 +1142,7 @@ static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
mlog_entry_void();
spin_lock_irqsave(&lockres->l_lock, flags);
lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
+ lockres_clear_flags(lockres, OCFS2_LOCK_UPCONVERT_FINISHING);
if (convert)
lockres->l_action = OCFS2_AST_INVALID;
else
@@ -1323,13 +1333,13 @@ static int __ocfs2_cluster_lock(struct ocfs2_super *osb,
again:
wait = 0;
+ spin_lock_irqsave(&lockres->l_lock, flags);
+
if (catch_signals && signal_pending(current)) {
ret = -ERESTARTSYS;
- goto out;
+ goto unlock;
}
- spin_lock_irqsave(&lockres->l_lock, flags);
-
mlog_bug_on_msg(lockres->l_flags & OCFS2_LOCK_FREEING,
"Cluster lock called on freeing lockres %s! flags "
"0x%lx\n", lockres->l_name, lockres->l_flags);
@@ -1346,6 +1356,25 @@ again:
goto unlock;
}
+ if (lockres->l_flags & OCFS2_LOCK_UPCONVERT_FINISHING) {
+ /*
+ * We've upconverted. If the lock now has a level we can
+ * work with, we take it. If, however, the lock is not at the
+ * required level, we go thru the full cycle. One way this could
+ * happen is if a process requesting an upconvert to PR is
+ * closely followed by another requesting upconvert to an EX.
+ * If the process requesting EX lands here, we want it to
+ * continue attempting to upconvert and let the process
+ * requesting PR take the lock.
+ * If multiple processes request upconvert to PR, the first one
+ * here will take the lock. The others will have to go thru the
+ * OCFS2_LOCK_BLOCKED check to ensure that there is no pending
+ * downconvert request.
+ */
+ if (level <= lockres->l_level)
+ goto update_holders;
+ }
+
if (lockres->l_flags & OCFS2_LOCK_BLOCKED &&
!ocfs2_may_continue_on_blocked_lock(lockres, level)) {
/* is the lock is currently blocked on behalf of
@@ -1416,11 +1445,14 @@ again:
goto again;
}
+update_holders:
/* Ok, if we get here then we're good to go. */
ocfs2_inc_holders(lockres, level);
ret = 0;
unlock:
+ lockres_clear_flags(lockres, OCFS2_LOCK_UPCONVERT_FINISHING);
+
spin_unlock_irqrestore(&lockres->l_lock, flags);
out:
/*
@@ -1855,7 +1887,7 @@ int ocfs2_file_lock(struct file *file, int ex, int trylock)
* outstanding lock request, so a cancel convert is
* required. We intentionally overwrite 'ret' - if the
* cancel fails and the lock was granted, it's easier
- * to just bubble sucess back up to the user.
+ * to just bubble success back up to the user.
*/
ret = ocfs2_flock_handle_signal(lockres, level);
} else if (!ret && (level > lockres->l_level)) {
@@ -3155,7 +3187,7 @@ out:
/* Mark the lockres as being dropped. It will no longer be
* queued if blocking, but we still may have to wait on it
* being dequeued from the downconvert thread before we can consider
- * it safe to drop.
+ * it safe to drop.
*
* You can *not* attempt to call cluster_lock on this lockres anymore. */
void ocfs2_mark_lockres_freeing(struct ocfs2_lock_res *lockres)
@@ -3352,6 +3384,7 @@ static int ocfs2_unblock_lock(struct ocfs2_super *osb,
unsigned long flags;
int blocking;
int new_level;
+ int level;
int ret = 0;
int set_lvb = 0;
unsigned int gen;
@@ -3360,9 +3393,17 @@ static int ocfs2_unblock_lock(struct ocfs2_super *osb,
spin_lock_irqsave(&lockres->l_lock, flags);
- BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
-
recheck:
+ /*
+ * Is it still blocking? If not, we have no more work to do.
+ */
+ if (!(lockres->l_flags & OCFS2_LOCK_BLOCKED)) {
+ BUG_ON(lockres->l_blocking != DLM_LOCK_NL);
+ spin_unlock_irqrestore(&lockres->l_lock, flags);
+ ret = 0;
+ goto leave;
+ }
+
if (lockres->l_flags & OCFS2_LOCK_BUSY) {
/* XXX
* This is a *big* race. The OCFS2_LOCK_PENDING flag
@@ -3401,6 +3442,31 @@ recheck:
goto leave;
}
+ /*
+ * This prevents livelocks. OCFS2_LOCK_UPCONVERT_FINISHING flag is
+ * set when the ast is received for an upconvert just before the
+ * OCFS2_LOCK_BUSY flag is cleared. Now if the fs received a bast
+ * on the heels of the ast, we want to delay the downconvert just
+ * enough to allow the up requestor to do its task. Because this
+ * lock is in the blocked queue, the lock will be downconverted
+ * as soon as the requestor is done with the lock.
+ */
+ if (lockres->l_flags & OCFS2_LOCK_UPCONVERT_FINISHING)
+ goto leave_requeue;
+
+ /*
+ * How can we block and yet be at NL? We were trying to upconvert
+ * from NL and got canceled. The code comes back here, and now
+ * we notice and clear BLOCKING.
+ */
+ if (lockres->l_level == DLM_LOCK_NL) {
+ BUG_ON(lockres->l_ex_holders || lockres->l_ro_holders);
+ lockres->l_blocking = DLM_LOCK_NL;
+ lockres_clear_flags(lockres, OCFS2_LOCK_BLOCKED);
+ spin_unlock_irqrestore(&lockres->l_lock, flags);
+ goto leave;
+ }
+
/* if we're blocking an exclusive and we have *any* holders,
* then requeue. */
if ((lockres->l_blocking == DLM_LOCK_EX)
@@ -3438,6 +3504,7 @@ recheck:
* may sleep, so we save off a copy of what we're blocking as
* it may change while we're not holding the spin lock. */
blocking = lockres->l_blocking;
+ level = lockres->l_level;
spin_unlock_irqrestore(&lockres->l_lock, flags);
ctl->unblock_action = lockres->l_ops->downconvert_worker(lockres, blocking);
@@ -3446,7 +3513,7 @@ recheck:
goto leave;
spin_lock_irqsave(&lockres->l_lock, flags);
- if (blocking != lockres->l_blocking) {
+ if ((blocking != lockres->l_blocking) || (level != lockres->l_level)) {
/* If this changed underneath us, then we can't drop
* it just yet. */
goto recheck;
diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c
index 15713cbb865..19ad145d2af 100644
--- a/fs/ocfs2/export.c
+++ b/fs/ocfs2/export.c
@@ -239,7 +239,7 @@ static int ocfs2_encode_fh(struct dentry *dentry, u32 *fh_in, int *max_len,
mlog(0, "Encoding parent: blkno: %llu, generation: %u\n",
(unsigned long long)blkno, generation);
}
-
+
*max_len = len;
bail:
diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c
index 843db64e9d4..5328529e7fd 100644
--- a/fs/ocfs2/extent_map.c
+++ b/fs/ocfs2/extent_map.c
@@ -37,6 +37,7 @@
#include "extent_map.h"
#include "inode.h"
#include "super.h"
+#include "symlink.h"
#include "buffer_head_io.h"
@@ -191,7 +192,7 @@ static int ocfs2_try_to_merge_extent_map(struct ocfs2_extent_map_item *emi,
emi->ei_clusters += ins->ei_clusters;
return 1;
} else if ((ins->ei_phys + ins->ei_clusters) == emi->ei_phys &&
- (ins->ei_cpos + ins->ei_clusters) == emi->ei_phys &&
+ (ins->ei_cpos + ins->ei_clusters) == emi->ei_cpos &&
ins->ei_flags == emi->ei_flags) {
emi->ei_phys = ins->ei_phys;
emi->ei_cpos = ins->ei_cpos;
@@ -703,6 +704,12 @@ out:
return ret;
}
+/*
+ * The ocfs2_fiemap_inline() may be a little bit misleading, since
+ * it not only handles the fiemap for inlined files, but also deals
+ * with the fast symlink, cause they have no difference for extent
+ * mapping per se.
+ */
static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh,
struct fiemap_extent_info *fieinfo,
u64 map_start)
@@ -715,11 +722,18 @@ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh,
struct ocfs2_inode_info *oi = OCFS2_I(inode);
di = (struct ocfs2_dinode *)di_bh->b_data;
- id_count = le16_to_cpu(di->id2.i_data.id_count);
+ if (ocfs2_inode_is_fast_symlink(inode))
+ id_count = ocfs2_fast_symlink_chars(inode->i_sb);
+ else
+ id_count = le16_to_cpu(di->id2.i_data.id_count);
if (map_start < id_count) {
phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits;
- phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data);
+ if (ocfs2_inode_is_fast_symlink(inode))
+ phys += offsetof(struct ocfs2_dinode, id2.i_symlink);
+ else
+ phys += offsetof(struct ocfs2_dinode,
+ id2.i_data.id_data);
ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count,
flags);
@@ -756,9 +770,10 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
down_read(&OCFS2_I(inode)->ip_alloc_sem);
/*
- * Handle inline-data separately.
+ * Handle inline-data and fast symlink separately.
*/
- if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
+ if ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) ||
+ ocfs2_inode_is_fast_symlink(inode)) {
ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start);
goto out_unlock;
}
@@ -786,6 +801,8 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
fe_flags = 0;
if (rec.e_flags & OCFS2_EXT_UNWRITTEN)
fe_flags |= FIEMAP_EXTENT_UNWRITTEN;
+ if (rec.e_flags & OCFS2_EXT_REFCOUNTED)
+ fe_flags |= FIEMAP_EXTENT_SHARED;
if (is_last)
fe_flags |= FIEMAP_EXTENT_LAST;
len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits;
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index de059f49058..558ce031242 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -749,7 +749,7 @@ static int ocfs2_write_zero_page(struct inode *inode,
int ret;
offset = (size & (PAGE_CACHE_SIZE-1)); /* Within page */
- /* ugh. in prepare/commit_write, if from==to==start of block, we
+ /* ugh. in prepare/commit_write, if from==to==start of block, we
** skip the prepare. make sure we never send an offset for the start
** of a block
*/
@@ -1772,13 +1772,14 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
loff_t *ppos,
size_t count,
int appending,
- int *direct_io)
+ int *direct_io,
+ int *has_refcount)
{
int ret = 0, meta_level = 0;
struct inode *inode = dentry->d_inode;
loff_t saved_pos, end;
- /*
+ /*
* We start with a read level meta lock and only jump to an ex
* if we need to make modifications here.
*/
@@ -1833,6 +1834,8 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
saved_pos,
count,
&meta_level);
+ if (has_refcount)
+ *has_refcount = 1;
}
if (ret < 0) {
@@ -1856,6 +1859,10 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
break;
}
+ if (has_refcount && *has_refcount == 1) {
+ *direct_io = 0;
+ break;
+ }
/*
* Allowing concurrent direct writes means
* i_size changes wouldn't be synchronized, so
@@ -1899,7 +1906,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
loff_t pos)
{
int ret, direct_io, appending, rw_level, have_alloc_sem = 0;
- int can_do_direct;
+ int can_do_direct, has_refcount = 0;
ssize_t written = 0;
size_t ocount; /* original count */
size_t count; /* after file limit checks */
@@ -1942,7 +1949,7 @@ relock:
can_do_direct = direct_io;
ret = ocfs2_prepare_inode_for_write(file->f_path.dentry, ppos,
iocb->ki_left, appending,
- &can_do_direct);
+ &can_do_direct, &has_refcount);
if (ret < 0) {
mlog_errno(ret);
goto out;
@@ -2006,14 +2013,16 @@ out_dio:
/* buffered aio wouldn't have proper lock coverage today */
BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
- if ((file->f_flags & O_SYNC && !direct_io) || IS_SYNC(inode)) {
+ if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) ||
+ ((file->f_flags & O_DIRECT) && has_refcount)) {
ret = filemap_fdatawrite_range(file->f_mapping, pos,
pos + count - 1);
if (ret < 0)
written = ret;
if (!ret && (old_size != i_size_read(inode) ||
- old_clusters != OCFS2_I(inode)->ip_clusters)) {
+ old_clusters != OCFS2_I(inode)->ip_clusters ||
+ has_refcount)) {
ret = jbd2_journal_force_commit(osb->journal->j_journal);
if (ret < 0)
written = ret;
@@ -2024,7 +2033,7 @@ out_dio:
pos + count - 1);
}
- /*
+ /*
* deep in g_f_a_w_n()->ocfs2_direct_IO we pass in a ocfs2_dio_end_io
* function pointer which is called when o_direct io completes so that
* it can unlock our rw lock. (it's the clustered equivalent of
@@ -2062,7 +2071,7 @@ static int ocfs2_splice_to_file(struct pipe_inode_info *pipe,
int ret;
ret = ocfs2_prepare_inode_for_write(out->f_path.dentry, &sd->pos,
- sd->total_len, 0, NULL);
+ sd->total_len, 0, NULL, NULL);
if (ret < 0) {
mlog_errno(ret);
return ret;
@@ -2189,7 +2198,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
goto bail;
}
- /*
+ /*
* buffered reads protect themselves in ->readpage(). O_DIRECT reads
* need locks to protect pending reads from racing with truncate.
*/
@@ -2211,10 +2220,10 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
* We're fine letting folks race truncates and extending
* writes with read across the cluster, just like they can
* locally. Hence no rw_lock during read.
- *
+ *
* Take and drop the meta data lock to update inode fields
* like i_size. This allows the checks down below
- * generic_file_aio_read() a chance of actually working.
+ * generic_file_aio_read() a chance of actually working.
*/
ret = ocfs2_inode_lock_atime(inode, filp->f_vfsmnt, &lock_level);
if (ret < 0) {
@@ -2239,7 +2248,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
bail:
if (have_alloc_sem)
up_read(&inode->i_alloc_sem);
- if (rw_level != -1)
+ if (rw_level != -1)
ocfs2_rw_unlock(inode, rw_level);
mlog_exit(ret);
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 0297fb8982b..88459bdd1ff 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -475,7 +475,7 @@ static int ocfs2_read_locked_inode(struct inode *inode,
if (args->fi_flags & OCFS2_FI_FLAG_ORPHAN_RECOVERY) {
status = ocfs2_try_open_lock(inode, 0);
if (status) {
- make_bad_inode(inode);
+ make_bad_inode(inode);
return status;
}
}
@@ -684,7 +684,7 @@ bail:
return status;
}
-/*
+/*
* Serialize with orphan dir recovery. If the process doing
* recovery on this orphan dir does an iget() with the dir
* i_mutex held, we'll deadlock here. Instead we detect this
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index 31fbb061951..7d9d9c132ce 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -7,6 +7,7 @@
#include <linux/fs.h>
#include <linux/mount.h>
+#include <linux/compat.h>
#define MLOG_MASK_PREFIX ML_INODE
#include <cluster/masklog.h>
@@ -181,6 +182,10 @@ long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
#ifdef CONFIG_COMPAT
long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
{
+ bool preserve;
+ struct reflink_arguments args;
+ struct inode *inode = file->f_path.dentry->d_inode;
+
switch (cmd) {
case OCFS2_IOC32_GETFLAGS:
cmd = OCFS2_IOC_GETFLAGS;
@@ -195,8 +200,15 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
case OCFS2_IOC_GROUP_EXTEND:
case OCFS2_IOC_GROUP_ADD:
case OCFS2_IOC_GROUP_ADD64:
- case OCFS2_IOC_REFLINK:
break;
+ case OCFS2_IOC_REFLINK:
+ if (copy_from_user(&args, (struct reflink_arguments *)arg,
+ sizeof(args)))
+ return -EFAULT;
+ preserve = (args.preserve != 0);
+
+ return ocfs2_reflink_ioctl(inode, compat_ptr(args.old_path),
+ compat_ptr(args.new_path), preserve);
default:
return -ENOIOCTLCMD;
}
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 54c16b66327..9336c60e3a3 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -659,7 +659,7 @@ static int __ocfs2_journal_access(handle_t *handle,
default:
status = -EINVAL;
- mlog(ML_ERROR, "Uknown access type!\n");
+ mlog(ML_ERROR, "Unknown access type!\n");
}
if (!status && ocfs2_meta_ecc(osb) && triggers)
jbd2_journal_set_triggers(bh, &triggers->ot_triggers);
@@ -2034,7 +2034,7 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb,
status = -ENOENT;
mlog_errno(status);
return status;
- }
+ }
mutex_lock(&orphan_dir_inode->i_mutex);
status = ocfs2_inode_lock(orphan_dir_inode, NULL, 0);
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index f010b22b1c4..50fb26a6a5f 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -2108,6 +2108,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
}
did_quota_inode = 1;
+ inode->i_nlink = 0;
/* do the real work now. */
status = ocfs2_mknod_locked(osb, dir, inode,
0, &new_di_bh, parent_di_bh, handle,
@@ -2136,6 +2137,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
if (status < 0)
mlog_errno(status);
+ insert_inode_hash(inode);
leave:
if (status < 0 && did_quota_inode)
vfs_dq_free_inode(inode);
@@ -2267,6 +2269,8 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
di = (struct ocfs2_dinode *)di_bh->b_data;
le32_add_cpu(&di->i_flags, -OCFS2_ORPHANED_FL);
di->i_orphaned_slot = 0;
+ inode->i_nlink = 1;
+ ocfs2_set_links_count(di, inode->i_nlink);
ocfs2_journal_dirty(handle, di_bh);
status = ocfs2_add_entry(handle, dentry, inode,
@@ -2284,7 +2288,6 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
goto out_commit;
}
- insert_inode_hash(inode);
dentry->d_op = &ocfs2_dentry_ops;
d_instantiate(dentry, inode);
status = 0;
@@ -2326,4 +2329,5 @@ const struct inode_operations ocfs2_dir_iops = {
.getxattr = generic_getxattr,
.listxattr = ocfs2_listxattr,
.removexattr = generic_removexattr,
+ .fiemap = ocfs2_fiemap,
};
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index d963d863870..740f448041e 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -136,6 +136,10 @@ enum ocfs2_unlock_action {
#define OCFS2_LOCK_PENDING (0x00000400) /* This lockres is pending a
call to dlm_lock. Only
exists with BUSY set. */
+#define OCFS2_LOCK_UPCONVERT_FINISHING (0x00000800) /* blocks the dc thread
+ * from downconverting
+ * before the upconvert
+ * has completed */
struct ocfs2_lock_res_ops;
@@ -245,9 +249,11 @@ enum ocfs2_mount_options
OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */
OCFS2_MOUNT_NOUSERXATTR = 1 << 6, /* No user xattr */
OCFS2_MOUNT_INODE64 = 1 << 7, /* Allow inode numbers > 2^32 */
- OCFS2_MOUNT_POSIX_ACL = 1 << 8, /* POSIX access control lists */
- OCFS2_MOUNT_USRQUOTA = 1 << 9, /* We support user quotas */
- OCFS2_MOUNT_GRPQUOTA = 1 << 10, /* We support group quotas */
+ OCFS2_MOUNT_POSIX_ACL = 1 << 8, /* Force POSIX access control lists */
+ OCFS2_MOUNT_NO_POSIX_ACL = 1 << 9, /* Disable POSIX access
+ control lists */
+ OCFS2_MOUNT_USRQUOTA = 1 << 10, /* We support user quotas */
+ OCFS2_MOUNT_GRPQUOTA = 1 << 11, /* We support group quotas */
};
#define OCFS2_OSB_SOFT_RO 0x0001
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index e9431e4a5e7..7638a38c32b 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -1202,7 +1202,7 @@ struct ocfs2_local_disk_dqinfo {
/* Header of one chunk of a quota file */
struct ocfs2_local_disk_chunk {
__le32 dqc_free; /* Number of free entries in the bitmap */
- u8 dqc_bitmap[0]; /* Bitmap of entries in the corresponding
+ __u8 dqc_bitmap[0]; /* Bitmap of entries in the corresponding
* chunk of quota file */
};
@@ -1417,9 +1417,16 @@ static inline int ocfs2_fast_symlink_chars(int blocksize)
return blocksize - offsetof(struct ocfs2_dinode, id2.i_symlink);
}
-static inline int ocfs2_max_inline_data(int blocksize)
+static inline int ocfs2_max_inline_data_with_xattr(int blocksize,
+ struct ocfs2_dinode *di)
{
- return blocksize - offsetof(struct ocfs2_dinode, id2.i_data.id_data);
+ if (di && (di->i_dyn_features & OCFS2_INLINE_XATTR_FL))
+ return blocksize -
+ offsetof(struct ocfs2_dinode, id2.i_data.id_data) -
+ di->i_xattr_inline_size;
+ else
+ return blocksize -
+ offsetof(struct ocfs2_dinode, id2.i_data.id_data);
}
static inline int ocfs2_extent_recs_per_inode(int blocksize)
diff --git a/fs/ocfs2/quota.h b/fs/ocfs2/quota.h
index e5df9d170b0..123bc520a2c 100644
--- a/fs/ocfs2/quota.h
+++ b/fs/ocfs2/quota.h
@@ -17,10 +17,6 @@
#include "ocfs2.h"
-/* Common stuff */
-/* id number of quota format */
-#define QFMT_OCFS2 3
-
/*
* In-memory structures
*/
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index 1a2c50a759f..21f9e71223c 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -1325,7 +1325,7 @@ out:
return status;
}
-static struct quota_format_ops ocfs2_format_ops = {
+static const struct quota_format_ops ocfs2_format_ops = {
.check_quota_file = ocfs2_local_check_quota_file,
.read_file_info = ocfs2_local_read_info,
.write_file_info = ocfs2_global_write_info,
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 3a0df7a1b81..8ae65c9c020 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -276,7 +276,7 @@ static void ocfs2_erase_refcount_tree_from_list(struct ocfs2_super *osb,
spin_unlock(&osb->osb_lock);
}
-void ocfs2_kref_remove_refcount_tree(struct kref *kref)
+static void ocfs2_kref_remove_refcount_tree(struct kref *kref)
{
struct ocfs2_refcount_tree *tree =
container_of(kref, struct ocfs2_refcount_tree, rf_getcnt);
@@ -524,23 +524,6 @@ out:
return ret;
}
-int ocfs2_lock_refcount_tree_by_inode(struct inode *inode, int rw,
- struct ocfs2_refcount_tree **ret_tree,
- struct buffer_head **ref_bh)
-{
- int ret;
- u64 ref_blkno;
-
- ret = ocfs2_get_refcount_block(inode, &ref_blkno);
- if (ret) {
- mlog_errno(ret);
- return ret;
- }
-
- return ocfs2_lock_refcount_tree(OCFS2_SB(inode->i_sb), ref_blkno,
- rw, ret_tree, ref_bh);
-}
-
void ocfs2_unlock_refcount_tree(struct ocfs2_super *osb,
struct ocfs2_refcount_tree *tree, int rw)
{
@@ -969,6 +952,103 @@ out:
}
/*
+ * Find the end range for a leaf refcount block indicated by
+ * el->l_recs[index].e_blkno.
+ */
+static int ocfs2_get_refcount_cpos_end(struct ocfs2_caching_info *ci,
+ struct buffer_head *ref_root_bh,
+ struct ocfs2_extent_block *eb,
+ struct ocfs2_extent_list *el,
+ int index, u32 *cpos_end)
+{
+ int ret, i, subtree_root;
+ u32 cpos;
+ u64 blkno;
+ struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
+ struct ocfs2_path *left_path = NULL, *right_path = NULL;
+ struct ocfs2_extent_tree et;
+ struct ocfs2_extent_list *tmp_el;
+
+ if (index < le16_to_cpu(el->l_next_free_rec) - 1) {
+ /*
+ * We have a extent rec after index, so just use the e_cpos
+ * of the next extent rec.
+ */
+ *cpos_end = le32_to_cpu(el->l_recs[index+1].e_cpos);
+ return 0;
+ }
+
+ if (!eb || (eb && !eb->h_next_leaf_blk)) {
+ /*
+ * We are the last extent rec, so any high cpos should
+ * be stored in this leaf refcount block.
+ */
+ *cpos_end = UINT_MAX;
+ return 0;
+ }
+
+ /*
+ * If the extent block isn't the last one, we have to find
+ * the subtree root between this extent block and the next
+ * leaf extent block and get the corresponding e_cpos from
+ * the subroot. Otherwise we may corrupt the b-tree.
+ */
+ ocfs2_init_refcount_extent_tree(&et, ci, ref_root_bh);
+
+ left_path = ocfs2_new_path_from_et(&et);
+ if (!left_path) {
+ ret = -ENOMEM;
+ mlog_errno(ret);
+ goto out;
+ }
+
+ cpos = le32_to_cpu(eb->h_list.l_recs[index].e_cpos);
+ ret = ocfs2_find_path(ci, left_path, cpos);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ right_path = ocfs2_new_path_from_path(left_path);
+ if (!right_path) {
+ ret = -ENOMEM;
+ mlog_errno(ret);
+ goto out;
+ }
+
+ ret = ocfs2_find_cpos_for_right_leaf(sb, left_path, &cpos);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ ret = ocfs2_find_path(ci, right_path, cpos);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ subtree_root = ocfs2_find_subtree_root(&et, left_path,
+ right_path);
+
+ tmp_el = left_path->p_node[subtree_root].el;
+ blkno = left_path->p_node[subtree_root+1].bh->b_blocknr;
+ for (i = 0; i < le32_to_cpu(tmp_el->l_next_free_rec); i++) {
+ if (le64_to_cpu(tmp_el->l_recs[i].e_blkno) == blkno) {
+ *cpos_end = le32_to_cpu(tmp_el->l_recs[i+1].e_cpos);
+ break;
+ }
+ }
+
+ BUG_ON(i == le32_to_cpu(tmp_el->l_next_free_rec));
+
+out:
+ ocfs2_free_path(left_path);
+ ocfs2_free_path(right_path);
+ return ret;
+}
+
+/*
* Given a cpos and len, try to find the refcount record which contains cpos.
* 1. If cpos can be found in one refcount record, return the record.
* 2. If cpos can't be found, return a fake record which start from cpos
@@ -983,10 +1063,10 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci,
struct buffer_head **ret_bh)
{
int ret = 0, i, found;
- u32 low_cpos;
+ u32 low_cpos, uninitialized_var(cpos_end);
struct ocfs2_extent_list *el;
- struct ocfs2_extent_rec *tmp, *rec = NULL;
- struct ocfs2_extent_block *eb;
+ struct ocfs2_extent_rec *rec = NULL;
+ struct ocfs2_extent_block *eb = NULL;
struct buffer_head *eb_bh = NULL, *ref_leaf_bh = NULL;
struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
struct ocfs2_refcount_block *rb =
@@ -1034,12 +1114,16 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci,
}
}
- /* adjust len when we have ocfs2_extent_rec after it. */
- if (found && i < le16_to_cpu(el->l_next_free_rec) - 1) {
- tmp = &el->l_recs[i+1];
+ if (found) {
+ ret = ocfs2_get_refcount_cpos_end(ci, ref_root_bh,
+ eb, el, i, &cpos_end);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
- if (le32_to_cpu(tmp->e_cpos) < cpos + len)
- len = le32_to_cpu(tmp->e_cpos) - cpos;
+ if (cpos_end < low_cpos + len)
+ len = cpos_end - low_cpos;
}
ret = ocfs2_read_refcount_block(ci, le64_to_cpu(rec->e_blkno),
@@ -1418,7 +1502,7 @@ static int ocfs2_divide_leaf_refcount_block(struct buffer_head *ref_leaf_bh,
/* change old and new rl_used accordingly. */
le16_add_cpu(&rl->rl_used, -num_moved);
- new_rl->rl_used = cpu_to_le32(num_moved);
+ new_rl->rl_used = cpu_to_le16(num_moved);
sort(&rl->rl_recs, le16_to_cpu(rl->rl_used),
sizeof(struct ocfs2_refcount_rec),
@@ -1797,7 +1881,8 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
recs_need++;
/* If the leaf block don't have enough record, expand it. */
- if (le16_to_cpu(rf_list->rl_used) + recs_need > rf_list->rl_count) {
+ if (le16_to_cpu(rf_list->rl_used) + recs_need >
+ le16_to_cpu(rf_list->rl_count)) {
struct ocfs2_refcount_rec tmp_rec;
u64 cpos = le64_to_cpu(orig_rec->r_cpos);
len = le32_to_cpu(orig_rec->r_clusters);
@@ -1859,7 +1944,7 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
memcpy(tail_rec, orig_rec, sizeof(struct ocfs2_refcount_rec));
le64_add_cpu(&tail_rec->r_cpos,
le32_to_cpu(tail_rec->r_clusters) - len);
- tail_rec->r_clusters = le32_to_cpu(len);
+ tail_rec->r_clusters = cpu_to_le32(len);
}
/*
@@ -2431,7 +2516,7 @@ out:
* we gonna touch and whether we need to create new blocks.
*
* Normally the refcount blocks store these refcount should be
- * continguous also, so that we can get the number easily.
+ * contiguous also, so that we can get the number easily.
* As for meta_ac, we will at most add split 2 refcount record and
* 2 more refcount block, so just check it in a rough way.
*
@@ -2860,7 +2945,7 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle,
while (offset < end) {
page_index = offset >> PAGE_CACHE_SHIFT;
- map_end = (page_index + 1) << PAGE_CACHE_SHIFT;
+ map_end = ((loff_t)page_index + 1) << PAGE_CACHE_SHIFT;
if (map_end > end)
map_end = end;
@@ -2872,8 +2957,12 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle,
page = grab_cache_page(mapping, page_index);
- /* This page can't be dirtied before we CoW it out. */
- BUG_ON(PageDirty(page));
+ /*
+ * In case PAGE_CACHE_SIZE <= CLUSTER_SIZE, This page
+ * can't be dirtied before we CoW it out.
+ */
+ if (PAGE_CACHE_SIZE <= OCFS2_SB(sb)->s_clustersize)
+ BUG_ON(PageDirty(page));
if (!PageUptodate(page)) {
ret = block_read_full_page(page, ocfs2_get_block);
@@ -3085,7 +3174,7 @@ static int ocfs2_cow_sync_writeback(struct super_block *sb,
while (offset < end) {
page_index = offset >> PAGE_CACHE_SHIFT;
- map_end = (page_index + 1) << PAGE_CACHE_SHIFT;
+ map_end = ((loff_t)page_index + 1) << PAGE_CACHE_SHIFT;
if (map_end > end)
map_end = end;
@@ -3840,8 +3929,7 @@ static int ocfs2_add_refcounted_extent(struct inode *inode,
}
ret = ocfs2_insert_extent(handle, et, cpos,
- cpu_to_le64(ocfs2_clusters_to_blocks(inode->i_sb,
- p_cluster)),
+ ocfs2_clusters_to_blocks(inode->i_sb, p_cluster),
num_clusters, ext_flags, meta_ac);
if (ret) {
mlog_errno(ret);
@@ -4253,8 +4341,8 @@ static int ocfs2_user_path_parent(const char __user *path,
* @new_dentry: target dentry
* @preserve: if true, preserve all file attributes
*/
-int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir,
- struct dentry *new_dentry, bool preserve)
+static int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir,
+ struct dentry *new_dentry, bool preserve)
{
struct inode *inode = old_dentry->d_inode;
int error;
diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index e49c4105026..3038c92af49 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -277,7 +277,7 @@ static int o2cb_cluster_connect(struct ocfs2_cluster_connection *conn)
u32 dlm_key;
struct dlm_ctxt *dlm;
struct o2dlm_private *priv;
- struct dlm_protocol_version dlm_version;
+ struct dlm_protocol_version fs_version;
BUG_ON(conn == NULL);
BUG_ON(o2cb_stack.sp_proto == NULL);
@@ -304,18 +304,18 @@ static int o2cb_cluster_connect(struct ocfs2_cluster_connection *conn)
/* used by the dlm code to make message headers unique, each
* node in this domain must agree on this. */
dlm_key = crc32_le(0, conn->cc_name, conn->cc_namelen);
- dlm_version.pv_major = conn->cc_version.pv_major;
- dlm_version.pv_minor = conn->cc_version.pv_minor;
+ fs_version.pv_major = conn->cc_version.pv_major;
+ fs_version.pv_minor = conn->cc_version.pv_minor;
- dlm = dlm_register_domain(conn->cc_name, dlm_key, &dlm_version);
+ dlm = dlm_register_domain(conn->cc_name, dlm_key, &fs_version);
if (IS_ERR(dlm)) {
rc = PTR_ERR(dlm);
mlog_errno(rc);
goto out_free;
}
- conn->cc_version.pv_major = dlm_version.pv_major;
- conn->cc_version.pv_minor = dlm_version.pv_minor;
+ conn->cc_version.pv_major = fs_version.pv_major;
+ conn->cc_version.pv_minor = fs_version.pv_minor;
conn->cc_lockspace = dlm;
dlm_register_eviction_cb(dlm, &priv->op_eviction_cb);
diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c
index ff4c798a563..da78a2a334f 100644
--- a/fs/ocfs2/stack_user.c
+++ b/fs/ocfs2/stack_user.c
@@ -814,7 +814,7 @@ static int fs_protocol_compare(struct ocfs2_protocol_version *existing,
static int user_cluster_connect(struct ocfs2_cluster_connection *conn)
{
dlm_lockspace_t *fsdlm;
- struct ocfs2_live_connection *control;
+ struct ocfs2_live_connection *uninitialized_var(control);
int rc = 0;
BUG_ON(conn == NULL);
diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c
index 3f2f1c45b7b..f3df0baa9a4 100644
--- a/fs/ocfs2/stackglue.c
+++ b/fs/ocfs2/stackglue.c
@@ -620,51 +620,46 @@ error:
static ctl_table ocfs2_nm_table[] = {
{
- .ctl_name = 1,
.procname = "hb_ctl_path",
.data = ocfs2_hb_ctl_path,
.maxlen = OCFS2_MAX_HB_CTL_PATH,
.mode = 0644,
- .proc_handler = &proc_dostring,
- .strategy = &sysctl_string,
+ .proc_handler = proc_dostring,
},
- { .ctl_name = 0 }
+ { }
};
static ctl_table ocfs2_mod_table[] = {
{
- .ctl_name = FS_OCFS2_NM,
.procname = "nm",
.data = NULL,
.maxlen = 0,
.mode = 0555,
.child = ocfs2_nm_table
},
- { .ctl_name = 0}
+ { }
};
static ctl_table ocfs2_kern_table[] = {
{
- .ctl_name = FS_OCFS2,
.procname = "ocfs2",
.data = NULL,
.maxlen = 0,
.mode = 0555,
.child = ocfs2_mod_table
},
- { .ctl_name = 0}
+ { }
};
static ctl_table ocfs2_root_table[] = {
{
- .ctl_name = CTL_FS,
.procname = "fs",
.data = NULL,
.maxlen = 0,
.mode = 0555,
.child = ocfs2_kern_table
},
- { .ctl_name = 0 }
+ { }
};
static struct ctl_table_header *ocfs2_table_header = NULL;
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 14f47d2bfe0..755cd49a5ef 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -100,6 +100,8 @@ struct mount_options
static int ocfs2_parse_options(struct super_block *sb, char *options,
struct mount_options *mopt,
int is_remount);
+static int ocfs2_check_set_options(struct super_block *sb,
+ struct mount_options *options);
static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt);
static void ocfs2_put_super(struct super_block *sb);
static int ocfs2_mount_volume(struct super_block *sb);
@@ -600,7 +602,8 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
lock_kernel();
- if (!ocfs2_parse_options(sb, data, &parsed_options, 1)) {
+ if (!ocfs2_parse_options(sb, data, &parsed_options, 1) ||
+ !ocfs2_check_set_options(sb, &parsed_options)) {
ret = -EINVAL;
goto out;
}
@@ -691,8 +694,6 @@ unlock_osb:
if (!ret) {
/* Only save off the new mount options in case of a successful
* remount. */
- if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
- parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
osb->s_mount_opt = parsed_options.mount_opt;
osb->s_atime_quantum = parsed_options.atime_quantum;
osb->preferred_slot = parsed_options.slot;
@@ -701,6 +702,10 @@ unlock_osb:
if (!ocfs2_is_hard_readonly(osb))
ocfs2_set_journal_params(osb);
+
+ sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+ ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ?
+ MS_POSIXACL : 0);
}
out:
unlock_kernel();
@@ -1011,31 +1016,16 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
brelse(bh);
bh = NULL;
- if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
- parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
-
+ if (!ocfs2_check_set_options(sb, &parsed_options)) {
+ status = -EINVAL;
+ goto read_super_error;
+ }
osb->s_mount_opt = parsed_options.mount_opt;
osb->s_atime_quantum = parsed_options.atime_quantum;
osb->preferred_slot = parsed_options.slot;
osb->osb_commit_interval = parsed_options.commit_interval;
osb->local_alloc_default_bits = ocfs2_megabytes_to_clusters(sb, parsed_options.localalloc_opt);
osb->local_alloc_bits = osb->local_alloc_default_bits;
- if (osb->s_mount_opt & OCFS2_MOUNT_USRQUOTA &&
- !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
- OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
- status = -EINVAL;
- mlog(ML_ERROR, "User quotas were requested, but this "
- "filesystem does not have the feature enabled.\n");
- goto read_super_error;
- }
- if (osb->s_mount_opt & OCFS2_MOUNT_GRPQUOTA &&
- !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
- status = -EINVAL;
- mlog(ML_ERROR, "Group quotas were requested, but this "
- "filesystem does not have the feature enabled.\n");
- goto read_super_error;
- }
status = ocfs2_verify_userspace_stack(osb, &parsed_options);
if (status)
@@ -1072,7 +1062,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
"file system, but write access is "
"unavailable.\n");
else
- mlog_errno(status);
+ mlog_errno(status);
goto read_super_error;
}
@@ -1245,6 +1235,40 @@ static struct file_system_type ocfs2_fs_type = {
.next = NULL
};
+static int ocfs2_check_set_options(struct super_block *sb,
+ struct mount_options *options)
+{
+ if (options->mount_opt & OCFS2_MOUNT_USRQUOTA &&
+ !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
+ mlog(ML_ERROR, "User quotas were requested, but this "
+ "filesystem does not have the feature enabled.\n");
+ return 0;
+ }
+ if (options->mount_opt & OCFS2_MOUNT_GRPQUOTA &&
+ !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
+ mlog(ML_ERROR, "Group quotas were requested, but this "
+ "filesystem does not have the feature enabled.\n");
+ return 0;
+ }
+ if (options->mount_opt & OCFS2_MOUNT_POSIX_ACL &&
+ !OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR)) {
+ mlog(ML_ERROR, "ACL support requested but extended attributes "
+ "feature is not enabled\n");
+ return 0;
+ }
+ /* No ACL setting specified? Use XATTR feature... */
+ if (!(options->mount_opt & (OCFS2_MOUNT_POSIX_ACL |
+ OCFS2_MOUNT_NO_POSIX_ACL))) {
+ if (OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR))
+ options->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+ else
+ options->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL;
+ }
+ return 1;
+}
+
static int ocfs2_parse_options(struct super_block *sb,
char *options,
struct mount_options *mopt,
@@ -1392,40 +1416,19 @@ static int ocfs2_parse_options(struct super_block *sb,
mopt->mount_opt |= OCFS2_MOUNT_INODE64;
break;
case Opt_usrquota:
- /* We check only on remount, otherwise features
- * aren't yet initialized. */
- if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
- OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
- mlog(ML_ERROR, "User quota requested but "
- "filesystem feature is not set\n");
- status = 0;
- goto bail;
- }
mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA;
break;
case Opt_grpquota:
- if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
- mlog(ML_ERROR, "Group quota requested but "
- "filesystem feature is not set\n");
- status = 0;
- goto bail;
- }
mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA;
break;
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
case Opt_acl:
mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+ mopt->mount_opt &= ~OCFS2_MOUNT_NO_POSIX_ACL;
break;
case Opt_noacl:
+ mopt->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL;
mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
break;
-#else
- case Opt_acl:
- case Opt_noacl:
- printk(KERN_INFO "ocfs2 (no)acl options not supported\n");
- break;
-#endif
default:
mlog(ML_ERROR,
"Unrecognized mount option \"%s\" "
@@ -1502,12 +1505,10 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
if (opts & OCFS2_MOUNT_INODE64)
seq_printf(s, ",inode64");
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
if (opts & OCFS2_MOUNT_POSIX_ACL)
seq_printf(s, ",acl");
else
seq_printf(s, ",noacl");
-#endif
return 0;
}
diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c
index e3421030a69..32499d213fc 100644
--- a/fs/ocfs2/symlink.c
+++ b/fs/ocfs2/symlink.c
@@ -137,20 +137,20 @@ static void *ocfs2_fast_follow_link(struct dentry *dentry,
}
memcpy(link, target, len);
- nd_set_link(nd, link);
bail:
+ nd_set_link(nd, status ? ERR_PTR(status) : link);
brelse(bh);
mlog_exit(status);
- return status ? ERR_PTR(status) : link;
+ return NULL;
}
static void ocfs2_fast_put_link(struct dentry *dentry, struct nameidata *nd, void *cookie)
{
- char *link = cookie;
-
- kfree(link);
+ char *link = nd_get_link(nd);
+ if (!IS_ERR(link))
+ kfree(link);
}
const struct inode_operations ocfs2_symlink_inode_operations = {
@@ -163,6 +163,7 @@ const struct inode_operations ocfs2_symlink_inode_operations = {
.getxattr = generic_getxattr,
.listxattr = ocfs2_listxattr,
.removexattr = generic_removexattr,
+ .fiemap = ocfs2_fiemap,
};
const struct inode_operations ocfs2_fast_symlink_inode_operations = {
.readlink = ocfs2_readlink,
@@ -174,4 +175,5 @@ const struct inode_operations ocfs2_fast_symlink_inode_operations = {
.getxattr = generic_getxattr,
.listxattr = ocfs2_listxattr,
.removexattr = generic_removexattr,
+ .fiemap = ocfs2_fiemap,
};
diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c
index c61369342a2..a0a120e82b9 100644
--- a/fs/ocfs2/uptodate.c
+++ b/fs/ocfs2/uptodate.c
@@ -267,8 +267,8 @@ static int ocfs2_buffer_cached(struct ocfs2_caching_info *ci,
}
/* Warning: even if it returns true, this does *not* guarantee that
- * the block is stored in our inode metadata cache.
- *
+ * the block is stored in our inode metadata cache.
+ *
* This can be called under lock_buffer()
*/
int ocfs2_buffer_uptodate(struct ocfs2_caching_info *ci,
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index fe3419068df..8fc6fb071c6 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -98,10 +98,8 @@ static struct ocfs2_xattr_def_value_root def_xv = {
struct xattr_handler *ocfs2_xattr_handlers[] = {
&ocfs2_xattr_user_handler,
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
&ocfs2_xattr_acl_access_handler,
&ocfs2_xattr_acl_default_handler,
-#endif
&ocfs2_xattr_trusted_handler,
&ocfs2_xattr_security_handler,
NULL
@@ -109,12 +107,10 @@ struct xattr_handler *ocfs2_xattr_handlers[] = {
static struct xattr_handler *ocfs2_xattr_handler_map[OCFS2_XATTR_MAX] = {
[OCFS2_XATTR_INDEX_USER] = &ocfs2_xattr_user_handler,
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
[OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS]
= &ocfs2_xattr_acl_access_handler,
[OCFS2_XATTR_INDEX_POSIX_ACL_DEFAULT]
= &ocfs2_xattr_acl_default_handler,
-#endif
[OCFS2_XATTR_INDEX_TRUSTED] = &ocfs2_xattr_trusted_handler,
[OCFS2_XATTR_INDEX_SECURITY] = &ocfs2_xattr_security_handler,
};
@@ -205,8 +201,6 @@ static int ocfs2_get_xattr_tree_value_root(struct super_block *sb,
int offset,
struct ocfs2_xattr_value_root **xv,
struct buffer_head **bh);
-static int ocfs2_xattr_security_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags);
static inline u16 ocfs2_xattr_buckets_per_cluster(struct ocfs2_super *osb)
{
@@ -6066,7 +6060,7 @@ static int ocfs2_value_metas_in_xattr_header(struct super_block *sb,
* to the extent block, so just calculate a maximum record num.
*/
if (!xv->xr_list.l_tree_depth)
- *num_recs += xv->xr_list.l_next_free_rec;
+ *num_recs += le16_to_cpu(xv->xr_list.l_next_free_rec);
else
*num_recs += ocfs2_clusters_for_bytes(sb,
XATTR_SIZE_MAX);
@@ -6978,9 +6972,9 @@ int ocfs2_init_security_and_acl(struct inode *dir,
ret = ocfs2_init_security_get(inode, dir, &si);
if (!ret) {
- ret = ocfs2_xattr_security_set(inode, si.name,
- si.value, si.value_len,
- XATTR_CREATE);
+ ret = ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_SECURITY,
+ si.name, si.value, si.value_len,
+ XATTR_CREATE);
if (ret) {
mlog_errno(ret);
goto leave;
@@ -7008,9 +7002,9 @@ leave:
/*
* 'security' attributes support
*/
-static size_t ocfs2_xattr_security_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_security_list(struct dentry *dentry, char *list,
size_t list_size, const char *name,
- size_t name_len)
+ size_t name_len, int type)
{
const size_t prefix_len = XATTR_SECURITY_PREFIX_LEN;
const size_t total_len = prefix_len + name_len + 1;
@@ -7023,23 +7017,23 @@ static size_t ocfs2_xattr_security_list(struct inode *inode, char *list,
return total_len;
}
-static int ocfs2_xattr_security_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
+static int ocfs2_xattr_security_get(struct dentry *dentry, const char *name,
+ void *buffer, size_t size, int type)
{
if (strcmp(name, "") == 0)
return -EINVAL;
- return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_SECURITY, name,
- buffer, size);
+ return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_SECURITY,
+ name, buffer, size);
}
-static int ocfs2_xattr_security_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
+static int ocfs2_xattr_security_set(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags, int type)
{
if (strcmp(name, "") == 0)
return -EINVAL;
- return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_SECURITY, name, value,
- size, flags);
+ return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_SECURITY,
+ name, value, size, flags);
}
int ocfs2_init_security_get(struct inode *inode,
@@ -7076,9 +7070,9 @@ struct xattr_handler ocfs2_xattr_security_handler = {
/*
* 'trusted' attributes support
*/
-static size_t ocfs2_xattr_trusted_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_trusted_list(struct dentry *dentry, char *list,
size_t list_size, const char *name,
- size_t name_len)
+ size_t name_len, int type)
{
const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
const size_t total_len = prefix_len + name_len + 1;
@@ -7091,23 +7085,23 @@ static size_t ocfs2_xattr_trusted_list(struct inode *inode, char *list,
return total_len;
}
-static int ocfs2_xattr_trusted_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
+static int ocfs2_xattr_trusted_get(struct dentry *dentry, const char *name,
+ void *buffer, size_t size, int type)
{
if (strcmp(name, "") == 0)
return -EINVAL;
- return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_TRUSTED, name,
- buffer, size);
+ return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_TRUSTED,
+ name, buffer, size);
}
-static int ocfs2_xattr_trusted_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
+static int ocfs2_xattr_trusted_set(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags, int type)
{
if (strcmp(name, "") == 0)
return -EINVAL;
- return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_TRUSTED, name, value,
- size, flags);
+ return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_TRUSTED,
+ name, value, size, flags);
}
struct xattr_handler ocfs2_xattr_trusted_handler = {
@@ -7120,13 +7114,13 @@ struct xattr_handler ocfs2_xattr_trusted_handler = {
/*
* 'user' attributes support
*/
-static size_t ocfs2_xattr_user_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_user_list(struct dentry *dentry, char *list,
size_t list_size, const char *name,
- size_t name_len)
+ size_t name_len, int type)
{
const size_t prefix_len = XATTR_USER_PREFIX_LEN;
const size_t total_len = prefix_len + name_len + 1;
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
return 0;
@@ -7139,31 +7133,31 @@ static size_t ocfs2_xattr_user_list(struct inode *inode, char *list,
return total_len;
}
-static int ocfs2_xattr_user_get(struct inode *inode, const char *name,
- void *buffer, size_t size)
+static int ocfs2_xattr_user_get(struct dentry *dentry, const char *name,
+ void *buffer, size_t size, int type)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
if (strcmp(name, "") == 0)
return -EINVAL;
if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
return -EOPNOTSUPP;
- return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_USER, name,
+ return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_USER, name,
buffer, size);
}
-static int ocfs2_xattr_user_set(struct inode *inode, const char *name,
- const void *value, size_t size, int flags)
+static int ocfs2_xattr_user_set(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags, int type)
{
- struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
if (strcmp(name, "") == 0)
return -EINVAL;
if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
return -EOPNOTSUPP;
- return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_USER, name, value,
- size, flags);
+ return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_USER,
+ name, value, size, flags);
}
struct xattr_handler ocfs2_xattr_user_handler = {
diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h
index 08e36389f56..abd72a47f52 100644
--- a/fs/ocfs2/xattr.h
+++ b/fs/ocfs2/xattr.h
@@ -40,10 +40,8 @@ struct ocfs2_security_xattr_info {
extern struct xattr_handler ocfs2_xattr_user_handler;
extern struct xattr_handler ocfs2_xattr_trusted_handler;
extern struct xattr_handler ocfs2_xattr_security_handler;
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
extern struct xattr_handler ocfs2_xattr_acl_access_handler;
extern struct xattr_handler ocfs2_xattr_acl_default_handler;
-#endif
extern struct xattr_handler *ocfs2_xattr_handlers[];
ssize_t ocfs2_listxattr(struct dentry *, char *, size_t);