2005-12-15 15:31:24 -07:00
|
|
|
/* -*- mode: c; c-basic-offset: 8; -*-
|
|
|
|
* vim: noexpandtab sw=8 ts=8 sts=0:
|
|
|
|
*
|
|
|
|
* super.c
|
|
|
|
*
|
|
|
|
* load/unload driver, mount/dismount volumes
|
|
|
|
*
|
|
|
|
* Copyright (C) 2002, 2004 Oracle. All rights reserved.
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public
|
|
|
|
* License as published by the Free Software Foundation; either
|
|
|
|
* version 2 of the License, or (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public
|
|
|
|
* License along with this program; if not, write to the
|
|
|
|
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
|
|
* Boston, MA 021110-1307, USA.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/highmem.h>
|
|
|
|
#include <linux/utsname.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/random.h>
|
|
|
|
#include <linux/statfs.h>
|
|
|
|
#include <linux/moduleparam.h>
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/socket.h>
|
|
|
|
#include <linux/inet.h>
|
|
|
|
#include <linux/parser.h>
|
|
|
|
#include <linux/crc32.h>
|
|
|
|
#include <linux/debugfs.h>
|
2007-09-06 14:34:16 -06:00
|
|
|
#include <linux/mount.h>
|
2008-01-29 17:59:56 -07:00
|
|
|
#include <linux/seq_file.h>
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
#define MLOG_MASK_PREFIX ML_SUPER
|
|
|
|
#include <cluster/masklog.h>
|
|
|
|
|
|
|
|
#include "ocfs2.h"
|
|
|
|
|
|
|
|
/* this should be the only file to include a version 1 header */
|
|
|
|
#include "ocfs1_fs_compat.h"
|
|
|
|
|
|
|
|
#include "alloc.h"
|
|
|
|
#include "dlmglue.h"
|
|
|
|
#include "export.h"
|
|
|
|
#include "extent_map.h"
|
|
|
|
#include "heartbeat.h"
|
|
|
|
#include "inode.h"
|
|
|
|
#include "journal.h"
|
|
|
|
#include "localalloc.h"
|
|
|
|
#include "namei.h"
|
|
|
|
#include "slot_map.h"
|
|
|
|
#include "super.h"
|
|
|
|
#include "sysfile.h"
|
|
|
|
#include "uptodate.h"
|
|
|
|
#include "ver.h"
|
|
|
|
|
|
|
|
#include "buffer_head_io.h"
|
|
|
|
|
2006-12-06 21:33:20 -07:00
|
|
|
static struct kmem_cache *ocfs2_inode_cachep = NULL;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
/* OCFS2 needs to schedule several differnt types of work which
|
|
|
|
* require cluster locking, disk I/O, recovery waits, etc. Since these
|
|
|
|
* types of work tend to be heavy we avoid using the kernel events
|
|
|
|
* workqueue and schedule on our own. */
|
|
|
|
struct workqueue_struct *ocfs2_wq = NULL;
|
|
|
|
|
|
|
|
static struct dentry *ocfs2_debugfs_root = NULL;
|
|
|
|
|
|
|
|
MODULE_AUTHOR("Oracle");
|
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
|
2007-09-07 10:16:10 -06:00
|
|
|
struct mount_options
|
|
|
|
{
|
2007-11-07 15:40:36 -07:00
|
|
|
unsigned long commit_interval;
|
2007-09-07 10:16:10 -06:00
|
|
|
unsigned long mount_opt;
|
|
|
|
unsigned int atime_quantum;
|
|
|
|
signed short slot;
|
2007-12-20 15:58:11 -07:00
|
|
|
unsigned int localalloc_opt;
|
2007-09-07 10:16:10 -06:00
|
|
|
};
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
static int ocfs2_parse_options(struct super_block *sb, char *options,
|
2007-09-07 10:16:10 -06:00
|
|
|
struct mount_options *mopt,
|
2007-06-18 18:00:24 -06:00
|
|
|
int is_remount);
|
2007-09-06 14:34:16 -06:00
|
|
|
static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt);
|
2005-12-15 15:31:24 -07:00
|
|
|
static void ocfs2_put_super(struct super_block *sb);
|
|
|
|
static int ocfs2_mount_volume(struct super_block *sb);
|
|
|
|
static int ocfs2_remount(struct super_block *sb, int *flags, char *data);
|
|
|
|
static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err);
|
|
|
|
static int ocfs2_initialize_mem_caches(void);
|
|
|
|
static void ocfs2_free_mem_caches(void);
|
|
|
|
static void ocfs2_delete_osb(struct ocfs2_super *osb);
|
|
|
|
|
2006-06-23 03:02:58 -06:00
|
|
|
static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
static int ocfs2_sync_fs(struct super_block *sb, int wait);
|
|
|
|
|
|
|
|
static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb);
|
|
|
|
static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb);
|
2007-09-26 12:10:04 -06:00
|
|
|
static void ocfs2_release_system_inodes(struct ocfs2_super *osb);
|
2005-12-15 15:31:24 -07:00
|
|
|
static int ocfs2_check_volume(struct ocfs2_super *osb);
|
|
|
|
static int ocfs2_verify_volume(struct ocfs2_dinode *di,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
u32 sectsize);
|
|
|
|
static int ocfs2_initialize_super(struct super_block *sb,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
int sector_size);
|
|
|
|
static int ocfs2_get_sector(struct super_block *sb,
|
|
|
|
struct buffer_head **bh,
|
|
|
|
int block,
|
|
|
|
int sect_size);
|
|
|
|
static void ocfs2_write_super(struct super_block *sb);
|
|
|
|
static struct inode *ocfs2_alloc_inode(struct super_block *sb);
|
|
|
|
static void ocfs2_destroy_inode(struct inode *inode);
|
|
|
|
|
2007-02-12 01:55:41 -07:00
|
|
|
static const struct super_operations ocfs2_sops = {
|
2005-12-15 15:31:24 -07:00
|
|
|
.statfs = ocfs2_statfs,
|
|
|
|
.alloc_inode = ocfs2_alloc_inode,
|
|
|
|
.destroy_inode = ocfs2_destroy_inode,
|
|
|
|
.drop_inode = ocfs2_drop_inode,
|
|
|
|
.clear_inode = ocfs2_clear_inode,
|
|
|
|
.delete_inode = ocfs2_delete_inode,
|
|
|
|
.sync_fs = ocfs2_sync_fs,
|
|
|
|
.write_super = ocfs2_write_super,
|
|
|
|
.put_super = ocfs2_put_super,
|
|
|
|
.remount_fs = ocfs2_remount,
|
2007-09-06 14:34:16 -06:00
|
|
|
.show_options = ocfs2_show_options,
|
2005-12-15 15:31:24 -07:00
|
|
|
};
|
|
|
|
|
|
|
|
enum {
|
|
|
|
Opt_barrier,
|
|
|
|
Opt_err_panic,
|
|
|
|
Opt_err_ro,
|
|
|
|
Opt_intr,
|
|
|
|
Opt_nointr,
|
|
|
|
Opt_hb_none,
|
|
|
|
Opt_hb_local,
|
|
|
|
Opt_data_ordered,
|
|
|
|
Opt_data_writeback,
|
2006-11-15 00:48:42 -07:00
|
|
|
Opt_atime_quantum,
|
2007-06-18 18:00:24 -06:00
|
|
|
Opt_slot,
|
2007-11-07 15:40:36 -07:00
|
|
|
Opt_commit,
|
2007-12-20 15:58:11 -07:00
|
|
|
Opt_localalloc,
|
2007-12-20 17:49:04 -07:00
|
|
|
Opt_localflocks,
|
2005-12-15 15:31:24 -07:00
|
|
|
Opt_err,
|
|
|
|
};
|
|
|
|
|
|
|
|
static match_table_t tokens = {
|
|
|
|
{Opt_barrier, "barrier=%u"},
|
|
|
|
{Opt_err_panic, "errors=panic"},
|
|
|
|
{Opt_err_ro, "errors=remount-ro"},
|
|
|
|
{Opt_intr, "intr"},
|
|
|
|
{Opt_nointr, "nointr"},
|
|
|
|
{Opt_hb_none, OCFS2_HB_NONE},
|
|
|
|
{Opt_hb_local, OCFS2_HB_LOCAL},
|
|
|
|
{Opt_data_ordered, "data=ordered"},
|
|
|
|
{Opt_data_writeback, "data=writeback"},
|
2006-11-15 00:48:42 -07:00
|
|
|
{Opt_atime_quantum, "atime_quantum=%u"},
|
2007-06-18 18:00:24 -06:00
|
|
|
{Opt_slot, "preferred_slot=%u"},
|
2007-11-07 15:40:36 -07:00
|
|
|
{Opt_commit, "commit=%u"},
|
2007-12-20 15:58:11 -07:00
|
|
|
{Opt_localalloc, "localalloc=%d"},
|
2007-12-20 17:49:04 -07:00
|
|
|
{Opt_localflocks, "localflocks"},
|
2005-12-15 15:31:24 -07:00
|
|
|
{Opt_err, NULL}
|
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* write_super and sync_fs ripped right out of ext3.
|
|
|
|
*/
|
|
|
|
static void ocfs2_write_super(struct super_block *sb)
|
|
|
|
{
|
2006-01-09 16:59:25 -07:00
|
|
|
if (mutex_trylock(&sb->s_lock) != 0)
|
2005-12-15 15:31:24 -07:00
|
|
|
BUG();
|
|
|
|
sb->s_dirt = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_sync_fs(struct super_block *sb, int wait)
|
|
|
|
{
|
2007-09-26 12:10:04 -06:00
|
|
|
int status;
|
2005-12-15 15:31:24 -07:00
|
|
|
tid_t target;
|
|
|
|
struct ocfs2_super *osb = OCFS2_SB(sb);
|
|
|
|
|
|
|
|
sb->s_dirt = 0;
|
|
|
|
|
|
|
|
if (ocfs2_is_hard_readonly(osb))
|
|
|
|
return -EROFS;
|
|
|
|
|
|
|
|
if (wait) {
|
|
|
|
status = ocfs2_flush_truncate_log(osb);
|
|
|
|
if (status < 0)
|
|
|
|
mlog_errno(status);
|
|
|
|
} else {
|
|
|
|
ocfs2_schedule_truncate_log_flush(osb, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (journal_start_commit(OCFS2_SB(sb)->journal->j_journal, &target)) {
|
|
|
|
if (wait)
|
|
|
|
log_wait_commit(OCFS2_SB(sb)->journal->j_journal,
|
|
|
|
target);
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb)
|
|
|
|
{
|
|
|
|
struct inode *new = NULL;
|
|
|
|
int status = 0;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
2008-01-10 16:11:45 -07:00
|
|
|
new = ocfs2_iget(osb, osb->root_blkno, OCFS2_FI_FLAG_SYSFILE, 0);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (IS_ERR(new)) {
|
|
|
|
status = PTR_ERR(new);
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
osb->root_inode = new;
|
|
|
|
|
2008-01-10 16:11:45 -07:00
|
|
|
new = ocfs2_iget(osb, osb->system_dir_blkno, OCFS2_FI_FLAG_SYSFILE, 0);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (IS_ERR(new)) {
|
|
|
|
status = PTR_ERR(new);
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
osb->sys_root_inode = new;
|
|
|
|
|
|
|
|
for (i = OCFS2_FIRST_ONLINE_SYSTEM_INODE;
|
|
|
|
i <= OCFS2_LAST_GLOBAL_SYSTEM_INODE; i++) {
|
|
|
|
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
|
|
|
|
if (!new) {
|
|
|
|
ocfs2_release_system_inodes(osb);
|
|
|
|
status = -EINVAL;
|
|
|
|
mlog_errno(status);
|
|
|
|
/* FIXME: Should ERROR_RO_FS */
|
|
|
|
mlog(ML_ERROR, "Unable to load system inode %d, "
|
|
|
|
"possibly corrupt fs?", i);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
// the array now has one ref, so drop this one
|
|
|
|
iput(new);
|
|
|
|
}
|
|
|
|
|
|
|
|
bail:
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb)
|
|
|
|
{
|
|
|
|
struct inode *new = NULL;
|
|
|
|
int status = 0;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
for (i = OCFS2_LAST_GLOBAL_SYSTEM_INODE + 1;
|
|
|
|
i < NUM_SYSTEM_INODES;
|
|
|
|
i++) {
|
|
|
|
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
|
|
|
|
if (!new) {
|
|
|
|
ocfs2_release_system_inodes(osb);
|
|
|
|
status = -EINVAL;
|
|
|
|
mlog(ML_ERROR, "status=%d, sysfile=%d, slot=%d\n",
|
|
|
|
status, i, osb->slot_num);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
/* the array now has one ref, so drop this one */
|
|
|
|
iput(new);
|
|
|
|
}
|
|
|
|
|
|
|
|
bail:
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2007-09-26 12:10:04 -06:00
|
|
|
static void ocfs2_release_system_inodes(struct ocfs2_super *osb)
|
2005-12-15 15:31:24 -07:00
|
|
|
{
|
2007-09-26 12:10:04 -06:00
|
|
|
int i;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct inode *inode;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
for (i = 0; i < NUM_SYSTEM_INODES; i++) {
|
|
|
|
inode = osb->system_inodes[i];
|
|
|
|
if (inode) {
|
|
|
|
iput(inode);
|
|
|
|
osb->system_inodes[i] = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
inode = osb->sys_root_inode;
|
|
|
|
if (inode) {
|
|
|
|
iput(inode);
|
|
|
|
osb->sys_root_inode = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
inode = osb->root_inode;
|
|
|
|
if (inode) {
|
|
|
|
iput(inode);
|
|
|
|
osb->root_inode = NULL;
|
|
|
|
}
|
|
|
|
|
2007-09-26 12:10:04 -06:00
|
|
|
mlog_exit(0);
|
2005-12-15 15:31:24 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
/* We're allocating fs objects, use GFP_NOFS */
|
|
|
|
static struct inode *ocfs2_alloc_inode(struct super_block *sb)
|
|
|
|
{
|
|
|
|
struct ocfs2_inode_info *oi;
|
|
|
|
|
2006-12-06 21:33:14 -07:00
|
|
|
oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (!oi)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
return &oi->vfs_inode;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void ocfs2_destroy_inode(struct inode *inode)
|
|
|
|
{
|
|
|
|
kmem_cache_free(ocfs2_inode_cachep, OCFS2_I(inode));
|
|
|
|
}
|
|
|
|
|
2007-07-20 13:56:16 -06:00
|
|
|
static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
|
|
|
|
unsigned int cbits)
|
2005-12-15 15:31:24 -07:00
|
|
|
{
|
2007-07-20 13:56:16 -06:00
|
|
|
unsigned int bytes = 1 << cbits;
|
|
|
|
unsigned int trim = bytes;
|
|
|
|
unsigned int bitshift = 32;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* i_size and all block offsets in ocfs2 are always 64 bits
|
|
|
|
* wide. i_clusters is 32 bits, in cluster-sized units. So on
|
|
|
|
* 64 bit platforms, cluster size will be the limiting factor.
|
2005-12-15 15:31:24 -07:00
|
|
|
*/
|
|
|
|
|
|
|
|
#if BITS_PER_LONG == 32
|
|
|
|
# if defined(CONFIG_LBD)
|
2006-10-11 02:22:05 -06:00
|
|
|
BUILD_BUG_ON(sizeof(sector_t) != 8);
|
2007-07-20 13:56:16 -06:00
|
|
|
/*
|
|
|
|
* We might be limited by page cache size.
|
|
|
|
*/
|
|
|
|
if (bytes > PAGE_CACHE_SIZE) {
|
|
|
|
bytes = PAGE_CACHE_SIZE;
|
|
|
|
trim = 1;
|
|
|
|
/*
|
|
|
|
* Shift by 31 here so that we don't get larger than
|
|
|
|
* MAX_LFS_FILESIZE
|
|
|
|
*/
|
|
|
|
bitshift = 31;
|
|
|
|
}
|
2005-12-15 15:31:24 -07:00
|
|
|
# else
|
2007-07-20 13:56:16 -06:00
|
|
|
/*
|
|
|
|
* We are limited by the size of sector_t. Use block size, as
|
|
|
|
* that's what we expose to the VFS.
|
|
|
|
*/
|
|
|
|
bytes = 1 << bbits;
|
|
|
|
trim = 1;
|
|
|
|
bitshift = 31;
|
2005-12-15 15:31:24 -07:00
|
|
|
# endif
|
|
|
|
#endif
|
|
|
|
|
2007-07-20 13:56:16 -06:00
|
|
|
/*
|
|
|
|
* Trim by a whole cluster when we can actually approach the
|
|
|
|
* on-disk limits. Otherwise we can overflow i_clusters when
|
|
|
|
* an extent start is at the max offset.
|
|
|
|
*/
|
|
|
|
return (((unsigned long long)bytes) << bitshift) - trim;
|
2005-12-15 15:31:24 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
|
|
|
|
{
|
|
|
|
int incompat_features;
|
|
|
|
int ret = 0;
|
2007-09-07 10:16:10 -06:00
|
|
|
struct mount_options parsed_options;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct ocfs2_super *osb = OCFS2_SB(sb);
|
|
|
|
|
2007-09-07 10:16:10 -06:00
|
|
|
if (!ocfs2_parse_options(sb, data, &parsed_options, 1)) {
|
2005-12-15 15:31:24 -07:00
|
|
|
ret = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) !=
|
2007-09-07 10:16:10 -06:00
|
|
|
(parsed_options.mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
|
2005-12-15 15:31:24 -07:00
|
|
|
ret = -EINVAL;
|
|
|
|
mlog(ML_ERROR, "Cannot change heartbeat mode on remount\n");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK) !=
|
2007-09-07 10:16:10 -06:00
|
|
|
(parsed_options.mount_opt & OCFS2_MOUNT_DATA_WRITEBACK)) {
|
2005-12-15 15:31:24 -07:00
|
|
|
ret = -EINVAL;
|
|
|
|
mlog(ML_ERROR, "Cannot change data mode on remount\n");
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* We're going to/from readonly mode. */
|
|
|
|
if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
|
|
|
|
/* Lock here so the check of HARD_RO and the potential
|
|
|
|
* setting of SOFT_RO is atomic. */
|
|
|
|
spin_lock(&osb->osb_lock);
|
|
|
|
if (osb->osb_flags & OCFS2_OSB_HARD_RO) {
|
|
|
|
mlog(ML_ERROR, "Remount on readonly device is forbidden.\n");
|
|
|
|
ret = -EROFS;
|
|
|
|
goto unlock_osb;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (*flags & MS_RDONLY) {
|
|
|
|
mlog(0, "Going to ro mode.\n");
|
|
|
|
sb->s_flags |= MS_RDONLY;
|
|
|
|
osb->osb_flags |= OCFS2_OSB_SOFT_RO;
|
|
|
|
} else {
|
|
|
|
mlog(0, "Making ro filesystem writeable.\n");
|
|
|
|
|
|
|
|
if (osb->osb_flags & OCFS2_OSB_ERROR_FS) {
|
|
|
|
mlog(ML_ERROR, "Cannot remount RDWR "
|
|
|
|
"filesystem due to previous errors.\n");
|
|
|
|
ret = -EROFS;
|
|
|
|
goto unlock_osb;
|
|
|
|
}
|
|
|
|
incompat_features = OCFS2_HAS_RO_COMPAT_FEATURE(sb, ~OCFS2_FEATURE_RO_COMPAT_SUPP);
|
|
|
|
if (incompat_features) {
|
|
|
|
mlog(ML_ERROR, "Cannot remount RDWR because "
|
|
|
|
"of unsupported optional features "
|
|
|
|
"(%x).\n", incompat_features);
|
|
|
|
ret = -EINVAL;
|
|
|
|
goto unlock_osb;
|
|
|
|
}
|
|
|
|
sb->s_flags &= ~MS_RDONLY;
|
|
|
|
osb->osb_flags &= ~OCFS2_OSB_SOFT_RO;
|
|
|
|
}
|
|
|
|
unlock_osb:
|
|
|
|
spin_unlock(&osb->osb_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!ret) {
|
|
|
|
/* Only save off the new mount options in case of a successful
|
|
|
|
* remount. */
|
2007-09-07 10:16:10 -06:00
|
|
|
osb->s_mount_opt = parsed_options.mount_opt;
|
|
|
|
osb->s_atime_quantum = parsed_options.atime_quantum;
|
|
|
|
osb->preferred_slot = parsed_options.slot;
|
2007-11-07 15:40:36 -07:00
|
|
|
if (parsed_options.commit_interval)
|
|
|
|
osb->osb_commit_interval = parsed_options.commit_interval;
|
2007-11-07 15:21:45 -07:00
|
|
|
|
|
|
|
if (!ocfs2_is_hard_readonly(osb))
|
|
|
|
ocfs2_set_journal_params(osb);
|
2005-12-15 15:31:24 -07:00
|
|
|
}
|
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_sb_probe(struct super_block *sb,
|
|
|
|
struct buffer_head **bh,
|
|
|
|
int *sector_size)
|
|
|
|
{
|
2007-09-26 12:10:04 -06:00
|
|
|
int status, tmpstat;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct ocfs1_vol_disk_hdr *hdr;
|
|
|
|
struct ocfs2_dinode *di;
|
|
|
|
int blksize;
|
|
|
|
|
|
|
|
*bh = NULL;
|
|
|
|
|
|
|
|
/* may be > 512 */
|
|
|
|
*sector_size = bdev_hardsect_size(sb->s_bdev);
|
|
|
|
if (*sector_size > OCFS2_MAX_BLOCKSIZE) {
|
|
|
|
mlog(ML_ERROR, "Hardware sector size too large: %d (max=%d)\n",
|
|
|
|
*sector_size, OCFS2_MAX_BLOCKSIZE);
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Can this really happen? */
|
|
|
|
if (*sector_size < OCFS2_MIN_BLOCKSIZE)
|
|
|
|
*sector_size = OCFS2_MIN_BLOCKSIZE;
|
|
|
|
|
|
|
|
/* check block zero for old format */
|
|
|
|
status = ocfs2_get_sector(sb, bh, 0, *sector_size);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
hdr = (struct ocfs1_vol_disk_hdr *) (*bh)->b_data;
|
|
|
|
if (hdr->major_version == OCFS1_MAJOR_VERSION) {
|
|
|
|
mlog(ML_ERROR, "incompatible version: %u.%u\n",
|
|
|
|
hdr->major_version, hdr->minor_version);
|
|
|
|
status = -EINVAL;
|
|
|
|
}
|
|
|
|
if (memcmp(hdr->signature, OCFS1_VOLUME_SIGNATURE,
|
|
|
|
strlen(OCFS1_VOLUME_SIGNATURE)) == 0) {
|
|
|
|
mlog(ML_ERROR, "incompatible volume signature: %8s\n",
|
|
|
|
hdr->signature);
|
|
|
|
status = -EINVAL;
|
|
|
|
}
|
|
|
|
brelse(*bh);
|
|
|
|
*bh = NULL;
|
|
|
|
if (status < 0) {
|
|
|
|
mlog(ML_ERROR, "This is an ocfs v1 filesystem which must be "
|
|
|
|
"upgraded before mounting with ocfs v2\n");
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Now check at magic offset for 512, 1024, 2048, 4096
|
|
|
|
* blocksizes. 4096 is the maximum blocksize because it is
|
|
|
|
* the minimum clustersize.
|
|
|
|
*/
|
|
|
|
status = -EINVAL;
|
|
|
|
for (blksize = *sector_size;
|
|
|
|
blksize <= OCFS2_MAX_BLOCKSIZE;
|
|
|
|
blksize <<= 1) {
|
|
|
|
tmpstat = ocfs2_get_sector(sb, bh,
|
|
|
|
OCFS2_SUPER_BLOCK_BLKNO,
|
|
|
|
blksize);
|
|
|
|
if (tmpstat < 0) {
|
|
|
|
status = tmpstat;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
di = (struct ocfs2_dinode *) (*bh)->b_data;
|
|
|
|
status = ocfs2_verify_volume(di, *bh, blksize);
|
|
|
|
if (status >= 0)
|
|
|
|
goto bail;
|
|
|
|
brelse(*bh);
|
|
|
|
*bh = NULL;
|
|
|
|
if (status != -EAGAIN)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
bail:
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
static int ocfs2_verify_heartbeat(struct ocfs2_super *osb)
|
|
|
|
{
|
|
|
|
if (ocfs2_mount_local(osb)) {
|
|
|
|
if (osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) {
|
|
|
|
mlog(ML_ERROR, "Cannot heartbeat on a locally "
|
|
|
|
"mounted device.\n");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!(osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
|
|
|
|
if (!ocfs2_mount_local(osb) && !ocfs2_is_hard_readonly(osb)) {
|
|
|
|
mlog(ML_ERROR, "Heartbeat has to be started to mount "
|
|
|
|
"a read-write clustered device.\n");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
|
|
|
|
{
|
|
|
|
struct dentry *root;
|
|
|
|
int status, sector_size;
|
2007-09-07 10:16:10 -06:00
|
|
|
struct mount_options parsed_options;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct inode *inode = NULL;
|
|
|
|
struct ocfs2_super *osb = NULL;
|
|
|
|
struct buffer_head *bh = NULL;
|
2006-12-05 18:56:35 -07:00
|
|
|
char nodestr[8];
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
mlog_entry("%p, %p, %i", sb, data, silent);
|
|
|
|
|
2007-09-07 10:16:10 -06:00
|
|
|
if (!ocfs2_parse_options(sb, data, &parsed_options, 0)) {
|
2005-12-15 15:31:24 -07:00
|
|
|
status = -EINVAL;
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* probe for superblock */
|
|
|
|
status = ocfs2_sb_probe(sb, &bh, §or_size);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog(ML_ERROR, "superblock probe failed!\n");
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_initialize_super(sb, bh, sector_size);
|
|
|
|
osb = OCFS2_SB(sb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
brelse(bh);
|
|
|
|
bh = NULL;
|
2007-09-07 10:16:10 -06:00
|
|
|
osb->s_mount_opt = parsed_options.mount_opt;
|
|
|
|
osb->s_atime_quantum = parsed_options.atime_quantum;
|
|
|
|
osb->preferred_slot = parsed_options.slot;
|
2007-11-07 15:40:36 -07:00
|
|
|
osb->osb_commit_interval = parsed_options.commit_interval;
|
2007-12-20 15:58:11 -07:00
|
|
|
osb->local_alloc_size = parsed_options.localalloc_opt;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
sb->s_magic = OCFS2_SUPER_MAGIC;
|
|
|
|
|
|
|
|
/* Hard readonly mode only if: bdev_read_only, MS_RDONLY,
|
|
|
|
* heartbeat=none */
|
|
|
|
if (bdev_read_only(sb->s_bdev)) {
|
|
|
|
if (!(sb->s_flags & MS_RDONLY)) {
|
|
|
|
status = -EACCES;
|
|
|
|
mlog(ML_ERROR, "Readonly device detected but readonly "
|
|
|
|
"mount was not specified.\n");
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* You should not be able to start a local heartbeat
|
|
|
|
* on a readonly device. */
|
|
|
|
if (osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) {
|
|
|
|
status = -EROFS;
|
|
|
|
mlog(ML_ERROR, "Local heartbeat specified on readonly "
|
|
|
|
"device.\n");
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_check_journals_nolocks(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
if (status == -EROFS)
|
|
|
|
mlog(ML_ERROR, "Recovery required on readonly "
|
|
|
|
"file system, but write access is "
|
|
|
|
"unavailable.\n");
|
|
|
|
else
|
|
|
|
mlog_errno(status);
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
ocfs2_set_ro_flag(osb, 1);
|
|
|
|
|
|
|
|
printk(KERN_NOTICE "Readonly device detected. No cluster "
|
|
|
|
"services will be utilized for this mount. Recovery "
|
|
|
|
"will be skipped.\n");
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!ocfs2_is_hard_readonly(osb)) {
|
|
|
|
if (sb->s_flags & MS_RDONLY)
|
|
|
|
ocfs2_set_ro_flag(osb, 0);
|
|
|
|
}
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
status = ocfs2_verify_heartbeat(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
osb->osb_debug_root = debugfs_create_dir(osb->uuid_str,
|
|
|
|
ocfs2_debugfs_root);
|
|
|
|
if (!osb->osb_debug_root) {
|
|
|
|
status = -EINVAL;
|
|
|
|
mlog(ML_ERROR, "Unable to create per-mount debugfs root.\n");
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_mount_volume(sb);
|
|
|
|
if (osb->root_inode)
|
|
|
|
inode = igrab(osb->root_inode);
|
|
|
|
|
|
|
|
if (status < 0)
|
|
|
|
goto read_super_error;
|
|
|
|
|
|
|
|
if (!inode) {
|
|
|
|
status = -EIO;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
root = d_alloc_root(inode);
|
|
|
|
if (!root) {
|
|
|
|
status = -ENOMEM;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto read_super_error;
|
|
|
|
}
|
|
|
|
|
|
|
|
sb->s_root = root;
|
|
|
|
|
|
|
|
ocfs2_complete_mount_recovery(osb);
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
if (ocfs2_mount_local(osb))
|
|
|
|
snprintf(nodestr, sizeof(nodestr), "local");
|
|
|
|
else
|
2008-01-30 16:38:24 -07:00
|
|
|
snprintf(nodestr, sizeof(nodestr), "%u", osb->node_num);
|
2006-12-05 18:56:35 -07:00
|
|
|
|
|
|
|
printk(KERN_INFO "ocfs2: Mounting device (%s) on (node %s, slot %d) "
|
2006-04-27 17:41:31 -06:00
|
|
|
"with %s data mode.\n",
|
2006-12-05 18:56:35 -07:00
|
|
|
osb->dev_str, nodestr, osb->slot_num,
|
2005-12-15 15:31:24 -07:00
|
|
|
osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK ? "writeback" :
|
|
|
|
"ordered");
|
|
|
|
|
|
|
|
atomic_set(&osb->vol_state, VOLUME_MOUNTED);
|
|
|
|
wake_up(&osb->osb_mount_event);
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
|
|
|
|
read_super_error:
|
|
|
|
if (bh != NULL)
|
|
|
|
brelse(bh);
|
|
|
|
|
|
|
|
if (inode)
|
|
|
|
iput(inode);
|
|
|
|
|
|
|
|
if (osb) {
|
|
|
|
atomic_set(&osb->vol_state, VOLUME_DISABLED);
|
|
|
|
wake_up(&osb->osb_mount_event);
|
|
|
|
ocfs2_dismount_volume(sb, 1);
|
|
|
|
}
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
[PATCH] VFS: Permit filesystem to override root dentry on mount
Extend the get_sb() filesystem operation to take an extra argument that
permits the VFS to pass in the target vfsmount that defines the mountpoint.
The filesystem is then required to manually set the superblock and root dentry
pointers. For most filesystems, this should be done with simple_set_mnt()
which will set the superblock pointer and then set the root dentry to the
superblock's s_root (as per the old default behaviour).
The get_sb() op now returns an integer as there's now no need to return the
superblock pointer.
This patch permits a superblock to be implicitly shared amongst several mount
points, such as can be done with NFS to avoid potential inode aliasing. In
such a case, simple_set_mnt() would not be called, and instead the mnt_root
and mnt_sb would be set directly.
The patch also makes the following changes:
(*) the get_sb_*() convenience functions in the core kernel now take a vfsmount
pointer argument and return an integer, so most filesystems have to change
very little.
(*) If one of the convenience function is not used, then get_sb() should
normally call simple_set_mnt() to instantiate the vfsmount. This will
always return 0, and so can be tail-called from get_sb().
(*) generic_shutdown_super() now calls shrink_dcache_sb() to clean up the
dcache upon superblock destruction rather than shrink_dcache_anon().
This is required because the superblock may now have multiple trees that
aren't actually bound to s_root, but that still need to be cleaned up. The
currently called functions assume that the whole tree is rooted at s_root,
and that anonymous dentries are not the roots of trees which results in
dentries being left unculled.
However, with the way NFS superblock sharing are currently set to be
implemented, these assumptions are violated: the root of the filesystem is
simply a dummy dentry and inode (the real inode for '/' may well be
inaccessible), and all the vfsmounts are rooted on anonymous[*] dentries
with child trees.
[*] Anonymous until discovered from another tree.
(*) The documentation has been adjusted, including the additional bit of
changing ext2_* into foo_* in the documentation.
[akpm@osdl.org: convert ipath_fs, do other stuff]
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Nathan Scott <nathans@sgi.com>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 03:02:57 -06:00
|
|
|
static int ocfs2_get_sb(struct file_system_type *fs_type,
|
|
|
|
int flags,
|
|
|
|
const char *dev_name,
|
|
|
|
void *data,
|
|
|
|
struct vfsmount *mnt)
|
2005-12-15 15:31:24 -07:00
|
|
|
{
|
[PATCH] VFS: Permit filesystem to override root dentry on mount
Extend the get_sb() filesystem operation to take an extra argument that
permits the VFS to pass in the target vfsmount that defines the mountpoint.
The filesystem is then required to manually set the superblock and root dentry
pointers. For most filesystems, this should be done with simple_set_mnt()
which will set the superblock pointer and then set the root dentry to the
superblock's s_root (as per the old default behaviour).
The get_sb() op now returns an integer as there's now no need to return the
superblock pointer.
This patch permits a superblock to be implicitly shared amongst several mount
points, such as can be done with NFS to avoid potential inode aliasing. In
such a case, simple_set_mnt() would not be called, and instead the mnt_root
and mnt_sb would be set directly.
The patch also makes the following changes:
(*) the get_sb_*() convenience functions in the core kernel now take a vfsmount
pointer argument and return an integer, so most filesystems have to change
very little.
(*) If one of the convenience function is not used, then get_sb() should
normally call simple_set_mnt() to instantiate the vfsmount. This will
always return 0, and so can be tail-called from get_sb().
(*) generic_shutdown_super() now calls shrink_dcache_sb() to clean up the
dcache upon superblock destruction rather than shrink_dcache_anon().
This is required because the superblock may now have multiple trees that
aren't actually bound to s_root, but that still need to be cleaned up. The
currently called functions assume that the whole tree is rooted at s_root,
and that anonymous dentries are not the roots of trees which results in
dentries being left unculled.
However, with the way NFS superblock sharing are currently set to be
implemented, these assumptions are violated: the root of the filesystem is
simply a dummy dentry and inode (the real inode for '/' may well be
inaccessible), and all the vfsmounts are rooted on anonymous[*] dentries
with child trees.
[*] Anonymous until discovered from another tree.
(*) The documentation has been adjusted, including the additional bit of
changing ext2_* into foo_* in the documentation.
[akpm@osdl.org: convert ipath_fs, do other stuff]
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: Nathan Scott <nathans@sgi.com>
Cc: Roland Dreier <rolandd@cisco.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 03:02:57 -06:00
|
|
|
return get_sb_bdev(fs_type, flags, dev_name, data, ocfs2_fill_super,
|
|
|
|
mnt);
|
2005-12-15 15:31:24 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct file_system_type ocfs2_fs_type = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.name = "ocfs2",
|
|
|
|
.get_sb = ocfs2_get_sb, /* is this called when we mount
|
|
|
|
* the fs? */
|
|
|
|
.kill_sb = kill_block_super, /* set to the generic one
|
|
|
|
* right now, but do we
|
|
|
|
* need to change that? */
|
2006-09-08 15:22:54 -06:00
|
|
|
.fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE,
|
2005-12-15 15:31:24 -07:00
|
|
|
.next = NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static int ocfs2_parse_options(struct super_block *sb,
|
|
|
|
char *options,
|
2007-09-07 10:16:10 -06:00
|
|
|
struct mount_options *mopt,
|
2005-12-15 15:31:24 -07:00
|
|
|
int is_remount)
|
|
|
|
{
|
|
|
|
int status;
|
|
|
|
char *p;
|
|
|
|
|
|
|
|
mlog_entry("remount: %d, options: \"%s\"\n", is_remount,
|
|
|
|
options ? options : "(none)");
|
|
|
|
|
2007-11-07 15:40:36 -07:00
|
|
|
mopt->commit_interval = 0;
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt = 0;
|
|
|
|
mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
|
|
|
|
mopt->slot = OCFS2_INVALID_SLOT;
|
2007-12-20 15:58:11 -07:00
|
|
|
mopt->localalloc_opt = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
if (!options) {
|
|
|
|
status = 1;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
while ((p = strsep(&options, ",")) != NULL) {
|
|
|
|
int token, option;
|
|
|
|
substring_t args[MAX_OPT_ARGS];
|
|
|
|
|
|
|
|
if (!*p)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
token = match_token(p, tokens, args);
|
|
|
|
switch (token) {
|
|
|
|
case Opt_hb_local:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_HB_LOCAL;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_hb_none:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt &= ~OCFS2_MOUNT_HB_LOCAL;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_barrier:
|
|
|
|
if (match_int(&args[0], &option)) {
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (option)
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_BARRIER;
|
2005-12-15 15:31:24 -07:00
|
|
|
else
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt &= ~OCFS2_MOUNT_BARRIER;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_intr:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt &= ~OCFS2_MOUNT_NOINTR;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_nointr:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_NOINTR;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_err_panic:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_err_ro:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt &= ~OCFS2_MOUNT_ERRORS_PANIC;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_data_ordered:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt &= ~OCFS2_MOUNT_DATA_WRITEBACK;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
|
|
|
case Opt_data_writeback:
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_DATA_WRITEBACK;
|
2005-12-15 15:31:24 -07:00
|
|
|
break;
|
2006-11-15 00:48:42 -07:00
|
|
|
case Opt_atime_quantum:
|
|
|
|
if (match_int(&args[0], &option)) {
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (option >= 0)
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->atime_quantum = option;
|
2006-11-15 00:48:42 -07:00
|
|
|
break;
|
2007-06-18 18:00:24 -06:00
|
|
|
case Opt_slot:
|
|
|
|
option = 0;
|
|
|
|
if (match_int(&args[0], &option)) {
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (option)
|
2007-09-07 10:16:10 -06:00
|
|
|
mopt->slot = (s16)option;
|
2007-06-18 18:00:24 -06:00
|
|
|
break;
|
2007-11-07 15:40:36 -07:00
|
|
|
case Opt_commit:
|
|
|
|
option = 0;
|
|
|
|
if (match_int(&args[0], &option)) {
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (option < 0)
|
|
|
|
return 0;
|
|
|
|
if (option == 0)
|
|
|
|
option = JBD_DEFAULT_MAX_COMMIT_AGE;
|
|
|
|
mopt->commit_interval = HZ * option;
|
|
|
|
break;
|
2007-12-20 15:58:11 -07:00
|
|
|
case Opt_localalloc:
|
|
|
|
option = 0;
|
|
|
|
if (match_int(&args[0], &option)) {
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (option >= 0 && (option <= ocfs2_local_alloc_size(sb) * 8))
|
|
|
|
mopt->localalloc_opt = option;
|
|
|
|
break;
|
2007-12-20 17:49:04 -07:00
|
|
|
case Opt_localflocks:
|
|
|
|
/*
|
|
|
|
* Changing this during remount could race
|
|
|
|
* flock() requests, or "unbalance" existing
|
|
|
|
* ones (e.g., a lock is taken in one mode but
|
|
|
|
* dropped in the other). If users care enough
|
|
|
|
* to flip locking modes during remount, we
|
|
|
|
* could add a "local" flag to individual
|
|
|
|
* flock structures for proper tracking of
|
|
|
|
* state.
|
|
|
|
*/
|
|
|
|
if (!is_remount)
|
|
|
|
mopt->mount_opt |= OCFS2_MOUNT_LOCALFLOCKS;
|
|
|
|
break;
|
2005-12-15 15:31:24 -07:00
|
|
|
default:
|
|
|
|
mlog(ML_ERROR,
|
|
|
|
"Unrecognized mount option \"%s\" "
|
|
|
|
"or missing value\n", p);
|
|
|
|
status = 0;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
status = 1;
|
|
|
|
|
|
|
|
bail:
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2007-09-06 14:34:16 -06:00
|
|
|
static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
|
|
|
|
{
|
|
|
|
struct ocfs2_super *osb = OCFS2_SB(mnt->mnt_sb);
|
|
|
|
unsigned long opts = osb->s_mount_opt;
|
|
|
|
|
|
|
|
if (opts & OCFS2_MOUNT_HB_LOCAL)
|
|
|
|
seq_printf(s, ",_netdev,heartbeat=local");
|
|
|
|
else
|
|
|
|
seq_printf(s, ",heartbeat=none");
|
|
|
|
|
|
|
|
if (opts & OCFS2_MOUNT_NOINTR)
|
|
|
|
seq_printf(s, ",nointr");
|
|
|
|
|
|
|
|
if (opts & OCFS2_MOUNT_DATA_WRITEBACK)
|
|
|
|
seq_printf(s, ",data=writeback");
|
|
|
|
else
|
|
|
|
seq_printf(s, ",data=ordered");
|
|
|
|
|
|
|
|
if (opts & OCFS2_MOUNT_BARRIER)
|
|
|
|
seq_printf(s, ",barrier=1");
|
|
|
|
|
|
|
|
if (opts & OCFS2_MOUNT_ERRORS_PANIC)
|
|
|
|
seq_printf(s, ",errors=panic");
|
|
|
|
else
|
|
|
|
seq_printf(s, ",errors=remount-ro");
|
|
|
|
|
|
|
|
if (osb->preferred_slot != OCFS2_INVALID_SLOT)
|
|
|
|
seq_printf(s, ",preferred_slot=%d", osb->preferred_slot);
|
|
|
|
|
|
|
|
if (osb->s_atime_quantum != OCFS2_DEFAULT_ATIME_QUANTUM)
|
|
|
|
seq_printf(s, ",atime_quantum=%u", osb->s_atime_quantum);
|
|
|
|
|
2007-11-07 15:40:36 -07:00
|
|
|
if (osb->osb_commit_interval)
|
|
|
|
seq_printf(s, ",commit=%u",
|
|
|
|
(unsigned) (osb->osb_commit_interval / HZ));
|
|
|
|
|
2007-12-20 15:58:11 -07:00
|
|
|
if (osb->local_alloc_size != OCFS2_DEFAULT_LOCAL_ALLOC_SIZE)
|
|
|
|
seq_printf(s, ",localalloc=%d", osb->local_alloc_size);
|
|
|
|
|
2007-12-20 17:49:04 -07:00
|
|
|
if (opts & OCFS2_MOUNT_LOCALFLOCKS)
|
|
|
|
seq_printf(s, ",localflocks,");
|
|
|
|
|
2007-09-06 14:34:16 -06:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
static int __init ocfs2_init(void)
|
|
|
|
{
|
|
|
|
int status;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
ocfs2_print_version();
|
|
|
|
|
|
|
|
status = init_ocfs2_uptodate_cache();
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_initialize_mem_caches();
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
ocfs2_wq = create_singlethread_workqueue("ocfs2_wq");
|
|
|
|
if (!ocfs2_wq) {
|
|
|
|
status = -ENOMEM;
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
ocfs2_debugfs_root = debugfs_create_dir("ocfs2", NULL);
|
|
|
|
if (!ocfs2_debugfs_root) {
|
|
|
|
status = -EFAULT;
|
|
|
|
mlog(ML_ERROR, "Unable to create ocfs2 debugfs root.\n");
|
|
|
|
}
|
|
|
|
|
2008-01-30 17:58:36 -07:00
|
|
|
ocfs2_set_locking_protocol();
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
leave:
|
|
|
|
if (status < 0) {
|
|
|
|
ocfs2_free_mem_caches();
|
|
|
|
exit_ocfs2_uptodate_cache();
|
|
|
|
}
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
|
|
|
|
if (status >= 0) {
|
|
|
|
return register_filesystem(&ocfs2_fs_type);
|
|
|
|
} else
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit ocfs2_exit(void)
|
|
|
|
{
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
if (ocfs2_wq) {
|
|
|
|
flush_workqueue(ocfs2_wq);
|
|
|
|
destroy_workqueue(ocfs2_wq);
|
|
|
|
}
|
|
|
|
|
|
|
|
debugfs_remove(ocfs2_debugfs_root);
|
|
|
|
|
|
|
|
ocfs2_free_mem_caches();
|
|
|
|
|
|
|
|
unregister_filesystem(&ocfs2_fs_type);
|
|
|
|
|
|
|
|
exit_ocfs2_uptodate_cache();
|
|
|
|
|
|
|
|
mlog_exit_void();
|
|
|
|
}
|
|
|
|
|
|
|
|
static void ocfs2_put_super(struct super_block *sb)
|
|
|
|
{
|
|
|
|
mlog_entry("(0x%p)\n", sb);
|
|
|
|
|
|
|
|
ocfs2_sync_blockdev(sb);
|
|
|
|
ocfs2_dismount_volume(sb, 0);
|
|
|
|
|
|
|
|
mlog_exit_void();
|
|
|
|
}
|
|
|
|
|
2006-06-23 03:02:58 -06:00
|
|
|
static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
|
2005-12-15 15:31:24 -07:00
|
|
|
{
|
|
|
|
struct ocfs2_super *osb;
|
|
|
|
u32 numbits, freebits;
|
|
|
|
int status;
|
|
|
|
struct ocfs2_dinode *bm_lock;
|
|
|
|
struct buffer_head *bh = NULL;
|
|
|
|
struct inode *inode = NULL;
|
|
|
|
|
2006-06-23 03:02:58 -06:00
|
|
|
mlog_entry("(%p, %p)\n", dentry->d_sb, buf);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2006-06-23 03:02:58 -06:00
|
|
|
osb = OCFS2_SB(dentry->d_sb);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
inode = ocfs2_get_system_file_inode(osb,
|
|
|
|
GLOBAL_BITMAP_SYSTEM_INODE,
|
|
|
|
OCFS2_INVALID_SLOT);
|
|
|
|
if (!inode) {
|
|
|
|
mlog(ML_ERROR, "failed to get bitmap inode\n");
|
|
|
|
status = -EIO;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
2007-10-18 16:30:42 -06:00
|
|
|
status = ocfs2_inode_lock(inode, &bh, 0);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
bm_lock = (struct ocfs2_dinode *) bh->b_data;
|
|
|
|
|
|
|
|
numbits = le32_to_cpu(bm_lock->id1.bitmap1.i_total);
|
|
|
|
freebits = numbits - le32_to_cpu(bm_lock->id1.bitmap1.i_used);
|
|
|
|
|
|
|
|
buf->f_type = OCFS2_SUPER_MAGIC;
|
2006-06-23 03:02:58 -06:00
|
|
|
buf->f_bsize = dentry->d_sb->s_blocksize;
|
2005-12-15 15:31:24 -07:00
|
|
|
buf->f_namelen = OCFS2_MAX_FILENAME_LEN;
|
|
|
|
buf->f_blocks = ((sector_t) numbits) *
|
|
|
|
(osb->s_clustersize >> osb->sb->s_blocksize_bits);
|
|
|
|
buf->f_bfree = ((sector_t) freebits) *
|
|
|
|
(osb->s_clustersize >> osb->sb->s_blocksize_bits);
|
|
|
|
buf->f_bavail = buf->f_bfree;
|
|
|
|
buf->f_files = numbits;
|
|
|
|
buf->f_ffree = freebits;
|
|
|
|
|
|
|
|
brelse(bh);
|
|
|
|
|
2007-10-18 16:30:42 -06:00
|
|
|
ocfs2_inode_unlock(inode, 0);
|
2005-12-15 15:31:24 -07:00
|
|
|
status = 0;
|
|
|
|
bail:
|
|
|
|
if (inode)
|
|
|
|
iput(inode);
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
2007-10-17 00:25:51 -06:00
|
|
|
static void ocfs2_inode_init_once(struct kmem_cache *cachep, void *data)
|
2005-12-15 15:31:24 -07:00
|
|
|
{
|
|
|
|
struct ocfs2_inode_info *oi = data;
|
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
oi->ip_flags = 0;
|
|
|
|
oi->ip_open_count = 0;
|
|
|
|
spin_lock_init(&oi->ip_lock);
|
|
|
|
ocfs2_extent_map_init(&oi->vfs_inode);
|
|
|
|
INIT_LIST_HEAD(&oi->ip_io_markers);
|
|
|
|
oi->ip_created_trans = 0;
|
|
|
|
oi->ip_last_trans = 0;
|
|
|
|
oi->ip_dir_start_lookup = 0;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
init_rwsem(&oi->ip_alloc_sem);
|
|
|
|
mutex_init(&oi->ip_io_mutex);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
oi->ip_blkno = 0ULL;
|
|
|
|
oi->ip_clusters = 0;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
ocfs2_lock_res_init_once(&oi->ip_rw_lockres);
|
2007-10-18 16:30:42 -06:00
|
|
|
ocfs2_lock_res_init_once(&oi->ip_inode_lockres);
|
2007-05-16 23:10:57 -06:00
|
|
|
ocfs2_lock_res_init_once(&oi->ip_open_lockres);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
ocfs2_metadata_cache_init(&oi->vfs_inode);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-05-16 23:10:57 -06:00
|
|
|
inode_init_once(&oi->vfs_inode);
|
2005-12-15 15:31:24 -07:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_initialize_mem_caches(void)
|
|
|
|
{
|
|
|
|
ocfs2_inode_cachep = kmem_cache_create("ocfs2_inode_cache",
|
2006-03-24 04:16:06 -07:00
|
|
|
sizeof(struct ocfs2_inode_info),
|
|
|
|
0,
|
|
|
|
(SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
|
|
|
|
SLAB_MEM_SPREAD),
|
2007-07-19 19:11:58 -06:00
|
|
|
ocfs2_inode_init_once);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (!ocfs2_inode_cachep)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void ocfs2_free_mem_caches(void)
|
|
|
|
{
|
|
|
|
if (ocfs2_inode_cachep)
|
|
|
|
kmem_cache_destroy(ocfs2_inode_cachep);
|
|
|
|
|
|
|
|
ocfs2_inode_cachep = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_get_sector(struct super_block *sb,
|
|
|
|
struct buffer_head **bh,
|
|
|
|
int block,
|
|
|
|
int sect_size)
|
|
|
|
{
|
|
|
|
if (!sb_set_blocksize(sb, sect_size)) {
|
|
|
|
mlog(ML_ERROR, "unable to set blocksize\n");
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
|
|
|
|
*bh = sb_getblk(sb, block);
|
|
|
|
if (!*bh) {
|
|
|
|
mlog_errno(-EIO);
|
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
lock_buffer(*bh);
|
|
|
|
if (!buffer_dirty(*bh))
|
|
|
|
clear_buffer_uptodate(*bh);
|
|
|
|
unlock_buffer(*bh);
|
|
|
|
ll_rw_block(READ, 1, bh);
|
|
|
|
wait_on_buffer(*bh);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_mount_volume(struct super_block *sb)
|
|
|
|
{
|
|
|
|
int status = 0;
|
|
|
|
int unlock_super = 0;
|
|
|
|
struct ocfs2_super *osb = OCFS2_SB(sb);
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
if (ocfs2_is_hard_readonly(osb))
|
|
|
|
goto leave;
|
|
|
|
|
|
|
|
status = ocfs2_dlm_init(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_super_lock(osb, 1);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
unlock_super = 1;
|
|
|
|
|
|
|
|
/* This will load up the node map and add ourselves to it. */
|
|
|
|
status = ocfs2_find_slot(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* load all node-local system inodes */
|
|
|
|
status = ocfs2_init_local_system_inodes(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_check_volume(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
|
|
|
status = ocfs2_truncate_log_init(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto leave;
|
|
|
|
}
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
if (ocfs2_mount_local(osb))
|
|
|
|
goto leave;
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
leave:
|
|
|
|
if (unlock_super)
|
|
|
|
ocfs2_super_unlock(osb, 1);
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
|
|
|
|
{
|
2008-02-01 16:03:57 -07:00
|
|
|
int tmp, hangup_needed = 0;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct ocfs2_super *osb = NULL;
|
2006-12-05 18:56:35 -07:00
|
|
|
char nodestr[8];
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
mlog_entry("(0x%p)\n", sb);
|
|
|
|
|
|
|
|
BUG_ON(!sb);
|
|
|
|
osb = OCFS2_SB(sb);
|
|
|
|
BUG_ON(!osb);
|
|
|
|
|
|
|
|
ocfs2_shutdown_local_alloc(osb);
|
|
|
|
|
|
|
|
ocfs2_truncate_log_shutdown(osb);
|
|
|
|
|
2008-02-01 13:03:57 -07:00
|
|
|
/* This will disable recovery and flush any recovery work. */
|
|
|
|
ocfs2_recovery_exit(osb);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
ocfs2_journal_shutdown(osb);
|
|
|
|
|
|
|
|
ocfs2_sync_blockdev(sb);
|
|
|
|
|
2008-02-01 15:39:35 -07:00
|
|
|
/* No cluster connection means we've failed during mount, so skip
|
|
|
|
* all the steps which depended on that to complete. */
|
|
|
|
if (osb->cconn) {
|
2005-12-15 15:31:24 -07:00
|
|
|
tmp = ocfs2_super_lock(osb, 1);
|
|
|
|
if (tmp < 0) {
|
|
|
|
mlog_errno(tmp);
|
|
|
|
return;
|
|
|
|
}
|
2007-10-04 15:47:09 -06:00
|
|
|
}
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-10-04 15:47:09 -06:00
|
|
|
if (osb->slot_num != OCFS2_INVALID_SLOT)
|
|
|
|
ocfs2_put_slot(osb);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2008-02-01 15:39:35 -07:00
|
|
|
if (osb->cconn)
|
2005-12-15 15:31:24 -07:00
|
|
|
ocfs2_super_unlock(osb, 1);
|
|
|
|
|
|
|
|
ocfs2_release_system_inodes(osb);
|
|
|
|
|
2008-01-29 17:59:56 -07:00
|
|
|
/*
|
|
|
|
* If we're dismounting due to mount error, mount.ocfs2 will clean
|
|
|
|
* up heartbeat. If we're a local mount, there is no heartbeat.
|
|
|
|
* If we failed before we got a uuid_str yet, we can't stop
|
|
|
|
* heartbeat. Otherwise, do it.
|
|
|
|
*/
|
|
|
|
if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str)
|
2008-02-01 16:03:57 -07:00
|
|
|
hangup_needed = 1;
|
|
|
|
|
|
|
|
if (osb->cconn)
|
|
|
|
ocfs2_dlm_shutdown(osb, hangup_needed);
|
|
|
|
|
|
|
|
debugfs_remove(osb->osb_debug_root);
|
|
|
|
|
|
|
|
if (hangup_needed)
|
2008-01-29 17:59:56 -07:00
|
|
|
ocfs2_cluster_hangup(osb->uuid_str, strlen(osb->uuid_str));
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
atomic_set(&osb->vol_state, VOLUME_DISMOUNTED);
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
if (ocfs2_mount_local(osb))
|
|
|
|
snprintf(nodestr, sizeof(nodestr), "local");
|
|
|
|
else
|
2008-01-30 16:38:24 -07:00
|
|
|
snprintf(nodestr, sizeof(nodestr), "%u", osb->node_num);
|
2006-12-05 18:56:35 -07:00
|
|
|
|
|
|
|
printk(KERN_INFO "ocfs2: Unmounting device (%s) on (node %s)\n",
|
|
|
|
osb->dev_str, nodestr);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
ocfs2_delete_osb(osb);
|
|
|
|
kfree(osb);
|
|
|
|
sb->s_dev = 0;
|
|
|
|
sb->s_fs_info = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_setup_osb_uuid(struct ocfs2_super *osb, const unsigned char *uuid,
|
|
|
|
unsigned uuid_bytes)
|
|
|
|
{
|
|
|
|
int i, ret;
|
|
|
|
char *ptr;
|
|
|
|
|
|
|
|
BUG_ON(uuid_bytes != OCFS2_VOL_UUID_LEN);
|
|
|
|
|
2006-12-13 01:34:52 -07:00
|
|
|
osb->uuid_str = kzalloc(OCFS2_VOL_UUID_LEN * 2 + 1, GFP_KERNEL);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (osb->uuid_str == NULL)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
for (i = 0, ptr = osb->uuid_str; i < OCFS2_VOL_UUID_LEN; i++) {
|
|
|
|
/* print with null */
|
|
|
|
ret = snprintf(ptr, 3, "%02X", uuid[i]);
|
|
|
|
if (ret != 2) /* drop super cleans up */
|
|
|
|
return -EINVAL;
|
|
|
|
/* then only advance past the last char */
|
|
|
|
ptr += 2;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_initialize_super(struct super_block *sb,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
int sector_size)
|
|
|
|
{
|
2007-09-26 12:10:04 -06:00
|
|
|
int status;
|
2007-07-20 13:56:16 -06:00
|
|
|
int i, cbits, bbits;
|
|
|
|
struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct inode *inode = NULL;
|
|
|
|
struct ocfs2_journal *journal;
|
|
|
|
__le32 uuid_net_key;
|
|
|
|
struct ocfs2_super *osb;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
2006-12-13 01:34:52 -07:00
|
|
|
osb = kzalloc(sizeof(struct ocfs2_super), GFP_KERNEL);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (!osb) {
|
|
|
|
status = -ENOMEM;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
sb->s_fs_info = osb;
|
|
|
|
sb->s_op = &ocfs2_sops;
|
|
|
|
sb->s_export_op = &ocfs2_export_ops;
|
2007-08-09 17:52:30 -06:00
|
|
|
sb->s_time_gran = 1;
|
2005-12-15 15:31:24 -07:00
|
|
|
sb->s_flags |= MS_NOATIME;
|
|
|
|
/* this is needed to support O_LARGEFILE */
|
2007-07-20 13:56:16 -06:00
|
|
|
cbits = le32_to_cpu(di->id2.i_super.s_clustersize_bits);
|
|
|
|
bbits = le32_to_cpu(di->id2.i_super.s_blocksize_bits);
|
|
|
|
sb->s_maxbytes = ocfs2_max_file_offset(bbits, cbits);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
osb->sb = sb;
|
|
|
|
/* Save off for ocfs2_rw_direct */
|
|
|
|
osb->s_sectsize_bits = blksize_bits(sector_size);
|
2006-01-27 02:32:52 -07:00
|
|
|
BUG_ON(!osb->s_sectsize_bits);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2007-09-24 16:56:19 -06:00
|
|
|
spin_lock_init(&osb->dc_task_lock);
|
|
|
|
init_waitqueue_head(&osb->dc_event);
|
|
|
|
osb->dc_work_sequence = 0;
|
|
|
|
osb->dc_wake_sequence = 0;
|
2005-12-15 15:31:24 -07:00
|
|
|
INIT_LIST_HEAD(&osb->blocked_lock_list);
|
|
|
|
osb->blocked_lock_count = 0;
|
|
|
|
spin_lock_init(&osb->osb_lock);
|
|
|
|
|
|
|
|
atomic_set(&osb->alloc_stats.moves, 0);
|
|
|
|
atomic_set(&osb->alloc_stats.local_data, 0);
|
|
|
|
atomic_set(&osb->alloc_stats.bitmap_data, 0);
|
|
|
|
atomic_set(&osb->alloc_stats.bg_allocs, 0);
|
|
|
|
atomic_set(&osb->alloc_stats.bg_extends, 0);
|
|
|
|
|
|
|
|
ocfs2_init_node_maps(osb);
|
|
|
|
|
|
|
|
snprintf(osb->dev_str, sizeof(osb->dev_str), "%u,%u",
|
|
|
|
MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev));
|
|
|
|
|
2008-02-01 13:03:57 -07:00
|
|
|
status = ocfs2_recovery_init(osb);
|
|
|
|
if (status) {
|
|
|
|
mlog(ML_ERROR, "Unable to initialize recovery state\n");
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
init_waitqueue_head(&osb->checkpoint_event);
|
|
|
|
atomic_set(&osb->needs_checkpoint, 0);
|
|
|
|
|
2006-11-15 00:48:42 -07:00
|
|
|
osb->s_atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
osb->slot_num = OCFS2_INVALID_SLOT;
|
|
|
|
|
|
|
|
osb->local_alloc_state = OCFS2_LA_UNUSED;
|
|
|
|
osb->local_alloc_bh = NULL;
|
|
|
|
|
|
|
|
init_waitqueue_head(&osb->osb_mount_event);
|
|
|
|
|
|
|
|
osb->vol_label = kmalloc(OCFS2_MAX_VOL_LABEL_LEN, GFP_KERNEL);
|
|
|
|
if (!osb->vol_label) {
|
|
|
|
mlog(ML_ERROR, "unable to alloc vol label\n");
|
|
|
|
status = -ENOMEM;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
osb->max_slots = le16_to_cpu(di->id2.i_super.s_max_slots);
|
|
|
|
if (osb->max_slots > OCFS2_MAX_SLOTS || osb->max_slots == 0) {
|
|
|
|
mlog(ML_ERROR, "Invalid number of node slots (%u)\n",
|
|
|
|
osb->max_slots);
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
2006-04-27 17:41:31 -06:00
|
|
|
mlog(0, "max_slots for this device: %u\n", osb->max_slots);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2006-02-22 18:35:08 -07:00
|
|
|
init_waitqueue_head(&osb->osb_wipe_event);
|
|
|
|
osb->osb_orphan_wipes = kcalloc(osb->max_slots,
|
|
|
|
sizeof(*osb->osb_orphan_wipes),
|
|
|
|
GFP_KERNEL);
|
|
|
|
if (!osb->osb_orphan_wipes) {
|
|
|
|
status = -ENOMEM;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
osb->s_feature_compat =
|
|
|
|
le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_compat);
|
|
|
|
osb->s_feature_ro_compat =
|
|
|
|
le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_ro_compat);
|
|
|
|
osb->s_feature_incompat =
|
|
|
|
le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_incompat);
|
|
|
|
|
|
|
|
if ((i = OCFS2_HAS_INCOMPAT_FEATURE(osb->sb, ~OCFS2_FEATURE_INCOMPAT_SUPP))) {
|
|
|
|
mlog(ML_ERROR, "couldn't mount because of unsupported "
|
|
|
|
"optional features (%x).\n", i);
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
if (!(osb->sb->s_flags & MS_RDONLY) &&
|
|
|
|
(i = OCFS2_HAS_RO_COMPAT_FEATURE(osb->sb, ~OCFS2_FEATURE_RO_COMPAT_SUPP))) {
|
|
|
|
mlog(ML_ERROR, "couldn't mount RDWR because of "
|
|
|
|
"unsupported optional features (%x).\n", i);
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
get_random_bytes(&osb->s_next_generation, sizeof(u32));
|
|
|
|
|
|
|
|
/* FIXME
|
|
|
|
* This should be done in ocfs2_journal_init(), but unknown
|
|
|
|
* ordering issues will cause the filesystem to crash.
|
|
|
|
* If anyone wants to figure out what part of the code
|
|
|
|
* refers to osb->journal before ocfs2_journal_init() is run,
|
|
|
|
* be my guest.
|
|
|
|
*/
|
|
|
|
/* initialize our journal structure */
|
|
|
|
|
2006-12-13 01:34:52 -07:00
|
|
|
journal = kzalloc(sizeof(struct ocfs2_journal), GFP_KERNEL);
|
2005-12-15 15:31:24 -07:00
|
|
|
if (!journal) {
|
|
|
|
mlog(ML_ERROR, "unable to alloc journal\n");
|
|
|
|
status = -ENOMEM;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
osb->journal = journal;
|
|
|
|
journal->j_osb = osb;
|
|
|
|
|
|
|
|
atomic_set(&journal->j_num_trans, 0);
|
|
|
|
init_rwsem(&journal->j_trans_barrier);
|
|
|
|
init_waitqueue_head(&journal->j_checkpointed);
|
|
|
|
spin_lock_init(&journal->j_lock);
|
|
|
|
journal->j_trans_id = (unsigned long) 1;
|
|
|
|
INIT_LIST_HEAD(&journal->j_la_cleanups);
|
2006-11-22 07:57:56 -07:00
|
|
|
INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery);
|
2005-12-15 15:31:24 -07:00
|
|
|
journal->j_state = OCFS2_JOURNAL_FREE;
|
|
|
|
|
|
|
|
/* get some pseudo constants for clustersize bits */
|
|
|
|
osb->s_clustersize_bits =
|
|
|
|
le32_to_cpu(di->id2.i_super.s_clustersize_bits);
|
|
|
|
osb->s_clustersize = 1 << osb->s_clustersize_bits;
|
|
|
|
mlog(0, "clusterbits=%d\n", osb->s_clustersize_bits);
|
|
|
|
|
|
|
|
if (osb->s_clustersize < OCFS2_MIN_CLUSTERSIZE ||
|
|
|
|
osb->s_clustersize > OCFS2_MAX_CLUSTERSIZE) {
|
|
|
|
mlog(ML_ERROR, "Volume has invalid cluster size (%d)\n",
|
|
|
|
osb->s_clustersize);
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ocfs2_clusters_to_blocks(osb->sb, le32_to_cpu(di->i_clusters) - 1)
|
|
|
|
> (u32)~0UL) {
|
|
|
|
mlog(ML_ERROR, "Volume might try to write to blocks beyond "
|
|
|
|
"what jbd can address in 32 bits.\n");
|
|
|
|
status = -EINVAL;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ocfs2_setup_osb_uuid(osb, di->id2.i_super.s_uuid,
|
|
|
|
sizeof(di->id2.i_super.s_uuid))) {
|
|
|
|
mlog(ML_ERROR, "Out of memory trying to setup our uuid.\n");
|
|
|
|
status = -ENOMEM;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
2006-05-04 13:03:26 -06:00
|
|
|
memcpy(&uuid_net_key, di->id2.i_super.s_uuid, sizeof(uuid_net_key));
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
strncpy(osb->vol_label, di->id2.i_super.s_label, 63);
|
|
|
|
osb->vol_label[63] = '\0';
|
|
|
|
osb->root_blkno = le64_to_cpu(di->id2.i_super.s_root_blkno);
|
|
|
|
osb->system_dir_blkno = le64_to_cpu(di->id2.i_super.s_system_dir_blkno);
|
|
|
|
osb->first_cluster_group_blkno =
|
|
|
|
le64_to_cpu(di->id2.i_super.s_first_cluster_group);
|
|
|
|
osb->fs_generation = le32_to_cpu(di->i_fs_generation);
|
|
|
|
mlog(0, "vol_label: %s\n", osb->vol_label);
|
|
|
|
mlog(0, "uuid: %s\n", osb->uuid_str);
|
2006-03-03 11:24:33 -07:00
|
|
|
mlog(0, "root_blkno=%llu, system_dir_blkno=%llu\n",
|
|
|
|
(unsigned long long)osb->root_blkno,
|
|
|
|
(unsigned long long)osb->system_dir_blkno);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
osb->osb_dlm_debug = ocfs2_new_dlm_debug();
|
|
|
|
if (!osb->osb_dlm_debug) {
|
|
|
|
status = -ENOMEM;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
atomic_set(&osb->vol_state, VOLUME_INIT);
|
|
|
|
|
|
|
|
/* load root, system_dir, and all global system inodes */
|
|
|
|
status = ocfs2_init_global_system_inodes(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* global bitmap
|
|
|
|
*/
|
|
|
|
inode = ocfs2_get_system_file_inode(osb, GLOBAL_BITMAP_SYSTEM_INODE,
|
|
|
|
OCFS2_INVALID_SLOT);
|
|
|
|
if (!inode) {
|
|
|
|
status = -EINVAL;
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno;
|
|
|
|
iput(inode);
|
|
|
|
|
2007-12-18 00:46:10 -07:00
|
|
|
osb->bitmap_cpg = ocfs2_group_bitmap_size(sb) * 8;
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
status = ocfs2_init_slot_info(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
|
|
|
|
bail:
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* will return: -EAGAIN if it is ok to keep searching for superblocks
|
|
|
|
* -EINVAL if there is a bad superblock
|
|
|
|
* 0 on success
|
|
|
|
*/
|
|
|
|
static int ocfs2_verify_volume(struct ocfs2_dinode *di,
|
|
|
|
struct buffer_head *bh,
|
|
|
|
u32 blksz)
|
|
|
|
{
|
|
|
|
int status = -EAGAIN;
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
if (memcmp(di->i_signature, OCFS2_SUPER_BLOCK_SIGNATURE,
|
|
|
|
strlen(OCFS2_SUPER_BLOCK_SIGNATURE)) == 0) {
|
|
|
|
status = -EINVAL;
|
|
|
|
if ((1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits)) != blksz) {
|
|
|
|
mlog(ML_ERROR, "found superblock with incorrect block "
|
|
|
|
"size: found %u, should be %u\n",
|
|
|
|
1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits),
|
|
|
|
blksz);
|
|
|
|
} else if (le16_to_cpu(di->id2.i_super.s_major_rev_level) !=
|
|
|
|
OCFS2_MAJOR_REV_LEVEL ||
|
|
|
|
le16_to_cpu(di->id2.i_super.s_minor_rev_level) !=
|
|
|
|
OCFS2_MINOR_REV_LEVEL) {
|
|
|
|
mlog(ML_ERROR, "found superblock with bad version: "
|
|
|
|
"found %u.%u, should be %u.%u\n",
|
|
|
|
le16_to_cpu(di->id2.i_super.s_major_rev_level),
|
|
|
|
le16_to_cpu(di->id2.i_super.s_minor_rev_level),
|
|
|
|
OCFS2_MAJOR_REV_LEVEL,
|
|
|
|
OCFS2_MINOR_REV_LEVEL);
|
|
|
|
} else if (bh->b_blocknr != le64_to_cpu(di->i_blkno)) {
|
|
|
|
mlog(ML_ERROR, "bad block number on superblock: "
|
2006-03-03 11:24:33 -07:00
|
|
|
"found %llu, should be %llu\n",
|
2007-04-27 17:01:25 -06:00
|
|
|
(unsigned long long)le64_to_cpu(di->i_blkno),
|
2006-03-03 11:24:33 -07:00
|
|
|
(unsigned long long)bh->b_blocknr);
|
2005-12-15 15:31:24 -07:00
|
|
|
} else if (le32_to_cpu(di->id2.i_super.s_clustersize_bits) < 12 ||
|
|
|
|
le32_to_cpu(di->id2.i_super.s_clustersize_bits) > 20) {
|
|
|
|
mlog(ML_ERROR, "bad cluster size found: %u\n",
|
|
|
|
1 << le32_to_cpu(di->id2.i_super.s_clustersize_bits));
|
|
|
|
} else if (!le64_to_cpu(di->id2.i_super.s_root_blkno)) {
|
|
|
|
mlog(ML_ERROR, "bad root_blkno: 0\n");
|
|
|
|
} else if (!le64_to_cpu(di->id2.i_super.s_system_dir_blkno)) {
|
|
|
|
mlog(ML_ERROR, "bad system_dir_blkno: 0\n");
|
|
|
|
} else if (le16_to_cpu(di->id2.i_super.s_max_slots) > OCFS2_MAX_SLOTS) {
|
|
|
|
mlog(ML_ERROR,
|
|
|
|
"Superblock slots found greater than file system "
|
|
|
|
"maximum: found %u, max %u\n",
|
|
|
|
le16_to_cpu(di->id2.i_super.s_max_slots),
|
|
|
|
OCFS2_MAX_SLOTS);
|
|
|
|
} else {
|
|
|
|
/* found it! */
|
|
|
|
status = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ocfs2_check_volume(struct ocfs2_super *osb)
|
|
|
|
{
|
2007-09-26 12:10:04 -06:00
|
|
|
int status;
|
2005-12-15 15:31:24 -07:00
|
|
|
int dirty;
|
2006-12-05 18:56:35 -07:00
|
|
|
int local;
|
2005-12-15 15:31:24 -07:00
|
|
|
struct ocfs2_dinode *local_alloc = NULL; /* only used if we
|
|
|
|
* recover
|
|
|
|
* ourselves. */
|
|
|
|
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
/* Init our journal object. */
|
|
|
|
status = ocfs2_journal_init(osb->journal, &dirty);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog(ML_ERROR, "Could not initialize journal!\n");
|
|
|
|
goto finally;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* If the journal was unmounted cleanly then we don't want to
|
|
|
|
* recover anything. Otherwise, journal_load will do that
|
|
|
|
* dirty work for us :) */
|
|
|
|
if (!dirty) {
|
|
|
|
status = ocfs2_journal_wipe(osb->journal, 0);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto finally;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
mlog(ML_NOTICE, "File system was not unmounted cleanly, "
|
|
|
|
"recovering volume.\n");
|
|
|
|
}
|
|
|
|
|
2006-12-05 18:56:35 -07:00
|
|
|
local = ocfs2_mount_local(osb);
|
|
|
|
|
2005-12-15 15:31:24 -07:00
|
|
|
/* will play back anything left in the journal. */
|
2006-12-05 18:56:35 -07:00
|
|
|
ocfs2_journal_load(osb->journal, local);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
|
|
|
if (dirty) {
|
|
|
|
/* recover my local alloc if we didn't unmount cleanly. */
|
|
|
|
status = ocfs2_begin_local_alloc_recovery(osb,
|
|
|
|
osb->slot_num,
|
|
|
|
&local_alloc);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto finally;
|
|
|
|
}
|
|
|
|
/* we complete the recovery process after we've marked
|
|
|
|
* ourselves as mounted. */
|
|
|
|
}
|
|
|
|
|
|
|
|
mlog(0, "Journal loaded.\n");
|
|
|
|
|
|
|
|
status = ocfs2_load_local_alloc(osb);
|
|
|
|
if (status < 0) {
|
|
|
|
mlog_errno(status);
|
|
|
|
goto finally;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (dirty) {
|
|
|
|
/* Recovery will be completed after we've mounted the
|
|
|
|
* rest of the volume. */
|
|
|
|
osb->dirty = 1;
|
|
|
|
osb->local_alloc_copy = local_alloc;
|
|
|
|
local_alloc = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* go through each journal, trylock it and if you get the
|
|
|
|
* lock, and it's marked as dirty, set the bit in the recover
|
|
|
|
* map and launch a recovery thread for it. */
|
|
|
|
status = ocfs2_mark_dead_nodes(osb);
|
|
|
|
if (status < 0)
|
|
|
|
mlog_errno(status);
|
|
|
|
|
|
|
|
finally:
|
|
|
|
if (local_alloc)
|
|
|
|
kfree(local_alloc);
|
|
|
|
|
|
|
|
mlog_exit(status);
|
|
|
|
return status;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The routine gets called from dismount or close whenever a dismount on
|
|
|
|
* volume is requested and the osb open count becomes 1.
|
|
|
|
* It will remove the osb from the global list and also free up all the
|
|
|
|
* initialized resources and fileobject.
|
|
|
|
*/
|
|
|
|
static void ocfs2_delete_osb(struct ocfs2_super *osb)
|
|
|
|
{
|
|
|
|
mlog_entry_void();
|
|
|
|
|
|
|
|
/* This function assumes that the caller has the main osb resource */
|
|
|
|
|
2008-02-01 12:59:09 -07:00
|
|
|
ocfs2_free_slot_info(osb);
|
2005-12-15 15:31:24 -07:00
|
|
|
|
2006-02-22 18:35:08 -07:00
|
|
|
kfree(osb->osb_orphan_wipes);
|
2005-12-15 15:31:24 -07:00
|
|
|
/* FIXME
|
|
|
|
* This belongs in journal shutdown, but because we have to
|
|
|
|
* allocate osb->journal at the start of ocfs2_initalize_osb(),
|
|
|
|
* we free it here.
|
|
|
|
*/
|
|
|
|
kfree(osb->journal);
|
|
|
|
if (osb->local_alloc_copy)
|
|
|
|
kfree(osb->local_alloc_copy);
|
|
|
|
kfree(osb->uuid_str);
|
|
|
|
ocfs2_put_dlm_debug(osb->osb_dlm_debug);
|
|
|
|
memset(osb, 0, sizeof(struct ocfs2_super));
|
|
|
|
|
|
|
|
mlog_exit_void();
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Put OCFS2 into a readonly state, or (if the user specifies it),
|
|
|
|
* panic(). We do not support continue-on-error operation. */
|
|
|
|
static void ocfs2_handle_error(struct super_block *sb)
|
|
|
|
{
|
|
|
|
struct ocfs2_super *osb = OCFS2_SB(sb);
|
|
|
|
|
|
|
|
if (osb->s_mount_opt & OCFS2_MOUNT_ERRORS_PANIC)
|
|
|
|
panic("OCFS2: (device %s): panic forced after error\n",
|
|
|
|
sb->s_id);
|
|
|
|
|
|
|
|
ocfs2_set_osb_flag(osb, OCFS2_OSB_ERROR_FS);
|
|
|
|
|
|
|
|
if (sb->s_flags & MS_RDONLY &&
|
|
|
|
(ocfs2_is_soft_readonly(osb) ||
|
|
|
|
ocfs2_is_hard_readonly(osb)))
|
|
|
|
return;
|
|
|
|
|
|
|
|
printk(KERN_CRIT "File system is now read-only due to the potential "
|
|
|
|
"of on-disk corruption. Please run fsck.ocfs2 once the file "
|
|
|
|
"system is unmounted.\n");
|
|
|
|
sb->s_flags |= MS_RDONLY;
|
|
|
|
ocfs2_set_ro_flag(osb, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static char error_buf[1024];
|
|
|
|
|
|
|
|
void __ocfs2_error(struct super_block *sb,
|
|
|
|
const char *function,
|
|
|
|
const char *fmt, ...)
|
|
|
|
{
|
|
|
|
va_list args;
|
|
|
|
|
|
|
|
va_start(args, fmt);
|
2006-12-06 21:37:04 -07:00
|
|
|
vsnprintf(error_buf, sizeof(error_buf), fmt, args);
|
2005-12-15 15:31:24 -07:00
|
|
|
va_end(args);
|
|
|
|
|
|
|
|
/* Not using mlog here because we want to show the actual
|
|
|
|
* function the error came from. */
|
|
|
|
printk(KERN_CRIT "OCFS2: ERROR (device %s): %s: %s\n",
|
|
|
|
sb->s_id, function, error_buf);
|
|
|
|
|
|
|
|
ocfs2_handle_error(sb);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Handle critical errors. This is intentionally more drastic than
|
|
|
|
* ocfs2_handle_error, so we only use for things like journal errors,
|
|
|
|
* etc. */
|
|
|
|
void __ocfs2_abort(struct super_block* sb,
|
|
|
|
const char *function,
|
|
|
|
const char *fmt, ...)
|
|
|
|
{
|
|
|
|
va_list args;
|
|
|
|
|
|
|
|
va_start(args, fmt);
|
2006-12-06 21:37:04 -07:00
|
|
|
vsnprintf(error_buf, sizeof(error_buf), fmt, args);
|
2005-12-15 15:31:24 -07:00
|
|
|
va_end(args);
|
|
|
|
|
|
|
|
printk(KERN_CRIT "OCFS2: abort (device %s): %s: %s\n",
|
|
|
|
sb->s_id, function, error_buf);
|
|
|
|
|
|
|
|
/* We don't have the cluster support yet to go straight to
|
|
|
|
* hard readonly in here. Until then, we want to keep
|
|
|
|
* ocfs2_abort() so that we can at least mark critical
|
|
|
|
* errors.
|
|
|
|
*
|
|
|
|
* TODO: This should abort the journal and alert other nodes
|
|
|
|
* that our slot needs recovery. */
|
|
|
|
|
|
|
|
/* Force a panic(). This stinks, but it's better than letting
|
|
|
|
* things continue without having a proper hard readonly
|
|
|
|
* here. */
|
|
|
|
OCFS2_SB(sb)->s_mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
|
|
|
|
ocfs2_handle_error(sb);
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(ocfs2_init);
|
|
|
|
module_exit(ocfs2_exit);
|