summaryrefslogtreecommitdiff
path: root/fs/ocfs2/super.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ocfs2/super.c')
-rw-r--r--fs/ocfs2/super.c1332
1 files changed, 626 insertions, 706 deletions
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 854d80955bf8..2c7ba1480f7a 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1,26 +1,10 @@
-/* -*- mode: c; c-basic-offset: 8; -*-
- * vim: noexpandtab sw=8 ts=8 sts=0:
- *
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
* 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>
@@ -35,13 +19,13 @@
#include <linux/blkdev.h>
#include <linux/socket.h>
#include <linux/inet.h>
-#include <linux/parser.h>
+#include <linux/fs_parser.h>
+#include <linux/fs_context.h>
#include <linux/crc32.h>
#include <linux/debugfs.h>
-#include <linux/mount.h>
#include <linux/seq_file.h>
#include <linux/quotaops.h>
-#include <linux/cleancache.h>
+#include <linux/signal.h>
#define CREATE_TRACE_POINTS
#include "ocfs2_trace.h"
@@ -68,50 +52,43 @@
#include "super.h"
#include "sysfile.h"
#include "uptodate.h"
-#include "ver.h"
#include "xattr.h"
#include "quota.h"
#include "refcounttree.h"
#include "suballoc.h"
#include "buffer_head_io.h"
+#include "filecheck.h"
-static struct kmem_cache *ocfs2_inode_cachep = NULL;
+static struct kmem_cache *ocfs2_inode_cachep;
struct kmem_cache *ocfs2_dquot_cachep;
struct kmem_cache *ocfs2_qf_chunk_cachep;
-/* OCFS2 needs to schedule several different 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;
+static struct dentry *ocfs2_debugfs_root;
MODULE_AUTHOR("Oracle");
MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("OCFS2 cluster file system");
struct mount_options
{
unsigned long commit_interval;
unsigned long mount_opt;
unsigned int atime_quantum;
- signed short slot;
+ unsigned short slot;
int localalloc_opt;
unsigned int resv_level;
int dir_resv_level;
char cluster_stack[OCFS2_STACK_LABEL_LEN + 1];
+ bool user_stack;
};
-static int ocfs2_parse_options(struct super_block *sb, char *options,
- struct mount_options *mopt,
- int is_remount);
+static int ocfs2_parse_param(struct fs_context *fc, struct fs_parameter *param);
static int ocfs2_check_set_options(struct super_block *sb,
struct mount_options *options);
static int ocfs2_show_options(struct seq_file *s, struct dentry *root);
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);
@@ -138,36 +115,36 @@ static int ocfs2_get_sector(struct super_block *sb,
int block,
int sect_size);
static struct inode *ocfs2_alloc_inode(struct super_block *sb);
-static void ocfs2_destroy_inode(struct inode *inode);
+static void ocfs2_free_inode(struct inode *inode);
static int ocfs2_susp_quotas(struct ocfs2_super *osb, int unsuspend);
static int ocfs2_enable_quotas(struct ocfs2_super *osb);
static void ocfs2_disable_quotas(struct ocfs2_super *osb);
+static struct dquot __rcu **ocfs2_get_dquots(struct inode *inode)
+{
+ return OCFS2_I(inode)->i_dquot;
+}
+
static const struct super_operations ocfs2_sops = {
.statfs = ocfs2_statfs,
.alloc_inode = ocfs2_alloc_inode,
- .destroy_inode = ocfs2_destroy_inode,
- .drop_inode = ocfs2_drop_inode,
+ .free_inode = ocfs2_free_inode,
+ .drop_inode = inode_just_drop,
.evict_inode = ocfs2_evict_inode,
.sync_fs = ocfs2_sync_fs,
.put_super = ocfs2_put_super,
- .remount_fs = ocfs2_remount,
.show_options = ocfs2_show_options,
.quota_read = ocfs2_quota_read,
.quota_write = ocfs2_quota_write,
+ .get_dquots = ocfs2_get_dquots,
};
enum {
Opt_barrier,
- Opt_err_panic,
- Opt_err_ro,
+ Opt_errors,
Opt_intr,
- Opt_nointr,
- Opt_hb_none,
- Opt_hb_local,
- Opt_hb_global,
- Opt_data_ordered,
- Opt_data_writeback,
+ Opt_heartbeat,
+ Opt_data,
Opt_atime_quantum,
Opt_slot,
Opt_commit,
@@ -175,48 +152,64 @@ enum {
Opt_localflocks,
Opt_stack,
Opt_user_xattr,
- Opt_nouser_xattr,
Opt_inode64,
Opt_acl,
- Opt_noacl,
Opt_usrquota,
Opt_grpquota,
- Opt_coherency_buffered,
- Opt_coherency_full,
+ Opt_coherency,
Opt_resv_level,
Opt_dir_resv_level,
- Opt_err,
+ Opt_journal_async_commit,
+};
+
+static const struct constant_table ocfs2_param_errors[] = {
+ {"panic", OCFS2_MOUNT_ERRORS_PANIC},
+ {"remount-ro", OCFS2_MOUNT_ERRORS_ROFS},
+ {"continue", OCFS2_MOUNT_ERRORS_CONT},
+ {}
+};
+
+static const struct constant_table ocfs2_param_heartbeat[] = {
+ {"local", OCFS2_MOUNT_HB_LOCAL},
+ {"none", OCFS2_MOUNT_HB_NONE},
+ {"global", OCFS2_MOUNT_HB_GLOBAL},
+ {}
+};
+
+static const struct constant_table ocfs2_param_data[] = {
+ {"writeback", OCFS2_MOUNT_DATA_WRITEBACK},
+ {"ordered", 0},
+ {}
};
-static const 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_hb_global, OCFS2_HB_GLOBAL},
- {Opt_data_ordered, "data=ordered"},
- {Opt_data_writeback, "data=writeback"},
- {Opt_atime_quantum, "atime_quantum=%u"},
- {Opt_slot, "preferred_slot=%u"},
- {Opt_commit, "commit=%u"},
- {Opt_localalloc, "localalloc=%d"},
- {Opt_localflocks, "localflocks"},
- {Opt_stack, "cluster_stack=%s"},
- {Opt_user_xattr, "user_xattr"},
- {Opt_nouser_xattr, "nouser_xattr"},
- {Opt_inode64, "inode64"},
- {Opt_acl, "acl"},
- {Opt_noacl, "noacl"},
- {Opt_usrquota, "usrquota"},
- {Opt_grpquota, "grpquota"},
- {Opt_coherency_buffered, "coherency=buffered"},
- {Opt_coherency_full, "coherency=full"},
- {Opt_resv_level, "resv_level=%u"},
- {Opt_dir_resv_level, "dir_resv_level=%u"},
- {Opt_err, NULL}
+static const struct constant_table ocfs2_param_coherency[] = {
+ {"buffered", OCFS2_MOUNT_COHERENCY_BUFFERED},
+ {"full", 0},
+ {}
+};
+
+static const struct fs_parameter_spec ocfs2_param_spec[] = {
+ fsparam_u32 ("barrier", Opt_barrier),
+ fsparam_enum ("errors", Opt_errors, ocfs2_param_errors),
+ fsparam_flag_no ("intr", Opt_intr),
+ fsparam_enum ("heartbeat", Opt_heartbeat, ocfs2_param_heartbeat),
+ fsparam_enum ("data", Opt_data, ocfs2_param_data),
+ fsparam_u32 ("atime_quantum", Opt_atime_quantum),
+ fsparam_u32 ("preferred_slot", Opt_slot),
+ fsparam_u32 ("commit", Opt_commit),
+ fsparam_s32 ("localalloc", Opt_localalloc),
+ fsparam_flag ("localflocks", Opt_localflocks),
+ fsparam_string ("cluster_stack", Opt_stack),
+ fsparam_flag_no ("user_xattr", Opt_user_xattr),
+ fsparam_flag ("inode64", Opt_inode64),
+ fsparam_flag_no ("acl", Opt_acl),
+ fsparam_flag ("usrquota", Opt_usrquota),
+ fsparam_flag ("grpquota", Opt_grpquota),
+ fsparam_enum ("coherency", Opt_coherency, ocfs2_param_coherency),
+ fsparam_u32 ("resv_level", Opt_resv_level),
+ fsparam_u32 ("dir_resv_level", Opt_dir_resv_level),
+ fsparam_flag ("journal_async_commit", Opt_journal_async_commit),
+ {}
};
#ifdef CONFIG_DEBUG_FS
@@ -226,32 +219,33 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
struct ocfs2_recovery_map *rm = osb->recovery_map;
struct ocfs2_orphan_scan *os = &osb->osb_orphan_scan;
int i, out = 0;
+ unsigned long flags;
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Id: %-s Uuid: %-s Gen: 0x%X Label: %-s\n",
"Device", osb->dev_str, osb->uuid_str,
osb->fs_generation, osb->vol_label);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => State: %d Flags: 0x%lX\n", "Volume",
atomic_read(&osb->vol_state), osb->osb_flags);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Block: %lu Cluster: %d\n", "Sizes",
osb->sb->s_blocksize, osb->s_clustersize);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Compat: 0x%X Incompat: 0x%X "
"ROcompat: 0x%X\n",
"Features", osb->s_feature_compat,
osb->s_feature_incompat, osb->s_feature_ro_compat);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Opts: 0x%lX AtimeQuanta: %u\n", "Mount",
osb->s_mount_opt, osb->s_atime_quantum);
if (cconn) {
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Stack: %s Name: %*s "
"Version: %d.%d\n", "Cluster",
(*osb->osb_cluster_stack == '\0' ?
@@ -261,42 +255,42 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
cconn->cc_version.pv_minor);
}
- spin_lock(&osb->dc_task_lock);
- out += snprintf(buf + out, len - out,
+ spin_lock_irqsave(&osb->dc_task_lock, flags);
+ out += scnprintf(buf + out, len - out,
"%10s => Pid: %d Count: %lu WakeSeq: %lu "
"WorkSeq: %lu\n", "DownCnvt",
(osb->dc_task ? task_pid_nr(osb->dc_task) : -1),
osb->blocked_lock_count, osb->dc_wake_sequence,
osb->dc_work_sequence);
- spin_unlock(&osb->dc_task_lock);
+ spin_unlock_irqrestore(&osb->dc_task_lock, flags);
spin_lock(&osb->osb_lock);
- out += snprintf(buf + out, len - out, "%10s => Pid: %d Nodes:",
+ out += scnprintf(buf + out, len - out, "%10s => Pid: %d Nodes:",
"Recovery",
(osb->recovery_thread_task ?
task_pid_nr(osb->recovery_thread_task) : -1));
if (rm->rm_used == 0)
- out += snprintf(buf + out, len - out, " None\n");
+ out += scnprintf(buf + out, len - out, " None\n");
else {
for (i = 0; i < rm->rm_used; i++)
- out += snprintf(buf + out, len - out, " %d",
+ out += scnprintf(buf + out, len - out, " %d",
rm->rm_entries[i]);
- out += snprintf(buf + out, len - out, "\n");
+ out += scnprintf(buf + out, len - out, "\n");
}
spin_unlock(&osb->osb_lock);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => Pid: %d Interval: %lu\n", "Commit",
(osb->commit_task ? task_pid_nr(osb->commit_task) : -1),
osb->osb_commit_interval);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => State: %d TxnId: %lu NumTxns: %d\n",
"Journal", osb->journal->j_state,
osb->journal->j_trans_id,
atomic_read(&osb->journal->j_num_trans));
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => GlobalAllocs: %d LocalAllocs: %d "
"SubAllocs: %d LAWinMoves: %d SAExtends: %d\n",
"Stats",
@@ -306,7 +300,7 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
atomic_read(&osb->alloc_stats.moves),
atomic_read(&osb->alloc_stats.bg_extends));
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => State: %u Descriptor: %llu Size: %u bits "
"Default: %u bits\n",
"LocalAlloc", osb->local_alloc_state,
@@ -314,7 +308,7 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
osb->local_alloc_bits, osb->local_alloc_default_bits);
spin_lock(&osb->osb_lock);
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s => InodeSlot: %d StolenInodes: %d, "
"MetaSlot: %d StolenMeta: %d\n", "Steal",
osb->s_inode_steal_slot,
@@ -323,20 +317,20 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
atomic_read(&osb->s_num_meta_stolen));
spin_unlock(&osb->osb_lock);
- out += snprintf(buf + out, len - out, "OrphanScan => ");
- out += snprintf(buf + out, len - out, "Local: %u Global: %u ",
+ out += scnprintf(buf + out, len - out, "OrphanScan => ");
+ out += scnprintf(buf + out, len - out, "Local: %u Global: %u ",
os->os_count, os->os_seqno);
- out += snprintf(buf + out, len - out, " Last Scan: ");
+ out += scnprintf(buf + out, len - out, " Last Scan: ");
if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
- out += snprintf(buf + out, len - out, "Disabled\n");
+ out += scnprintf(buf + out, len - out, "Disabled\n");
else
- out += snprintf(buf + out, len - out, "%lu seconds ago\n",
- (get_seconds() - os->os_scantime.tv_sec));
+ out += scnprintf(buf + out, len - out, "%lu seconds ago\n",
+ (unsigned long)(ktime_get_seconds() - os->os_scantime));
- out += snprintf(buf + out, len - out, "%10s => %3s %10s\n",
+ out += scnprintf(buf + out, len - out, "%10s => %3s %10s\n",
"Slots", "Num", "RecoGen");
for (i = 0; i < osb->max_slots; ++i) {
- out += snprintf(buf + out, len - out,
+ out += scnprintf(buf + out, len - out,
"%10s %c %3d %10d\n",
" ",
(i == osb->slot_num ? '*' : ' '),
@@ -416,10 +410,10 @@ static int ocfs2_sync_fs(struct super_block *sb, int wait)
ocfs2_schedule_truncate_log_flush(osb, 0);
}
- if (jbd2_journal_start_commit(OCFS2_SB(sb)->journal->j_journal,
+ if (jbd2_journal_start_commit(osb->journal->j_journal,
&target)) {
if (wait)
- jbd2_log_wait_commit(OCFS2_SB(sb)->journal->j_journal,
+ jbd2_log_wait_commit(osb->journal->j_journal,
target);
}
return 0;
@@ -467,9 +461,8 @@ static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb)
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
if (!new) {
ocfs2_release_system_inodes(osb);
- status = -EINVAL;
+ status = ocfs2_is_soft_readonly(osb) ? -EROFS : -EINVAL;
mlog_errno(status);
- /* FIXME: Should ERROR_RO_FS */
mlog(ML_ERROR, "Unable to load system inode %d, "
"possibly corrupt fs?", i);
goto bail;
@@ -498,7 +491,7 @@ static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb)
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
if (!new) {
ocfs2_release_system_inodes(osb);
- status = -EINVAL;
+ status = ocfs2_is_soft_readonly(osb) ? -EROFS : -EINVAL;
mlog(ML_ERROR, "status=%d, sysfile=%d, slot=%d\n",
status, i, osb->slot_num);
goto bail;
@@ -557,25 +550,23 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb)
{
struct ocfs2_inode_info *oi;
- oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS);
+ oi = alloc_inode_sb(sb, ocfs2_inode_cachep, GFP_NOFS);
if (!oi)
return NULL;
+ oi->i_sync_tid = 0;
+ oi->i_datasync_tid = 0;
+ memset(&oi->i_dquot, 0, sizeof(oi->i_dquot));
+
jbd2_journal_init_jbd_inode(&oi->ip_jinode, &oi->vfs_inode);
return &oi->vfs_inode;
}
-static void ocfs2_i_callback(struct rcu_head *head)
+static void ocfs2_free_inode(struct inode *inode)
{
- struct inode *inode = container_of(head, struct inode, i_rcu);
kmem_cache_free(ocfs2_inode_cachep, OCFS2_I(inode));
}
-static void ocfs2_destroy_inode(struct inode *inode)
-{
- call_rcu(&inode->i_rcu, ocfs2_i_callback);
-}
-
static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
unsigned int cbits)
{
@@ -590,13 +581,12 @@ static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
*/
#if BITS_PER_LONG == 32
-# if defined(CONFIG_LBDAF)
BUILD_BUG_ON(sizeof(sector_t) != 8);
/*
* We might be limited by page cache size.
*/
- if (bytes > PAGE_CACHE_SIZE) {
- bytes = PAGE_CACHE_SIZE;
+ if (bytes > PAGE_SIZE) {
+ bytes = PAGE_SIZE;
trim = 1;
/*
* Shift by 31 here so that we don't get larger than
@@ -604,15 +594,6 @@ static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
*/
bitshift = 31;
}
-# else
- /*
- * 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;
-# endif
#endif
/*
@@ -623,30 +604,32 @@ static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
return (((unsigned long long)bytes) << bitshift) - trim;
}
-static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
+static int ocfs2_reconfigure(struct fs_context *fc)
{
int incompat_features;
int ret = 0;
- struct mount_options parsed_options;
+ struct mount_options *parsed_options = fc->fs_private;
+ struct super_block *sb = fc->root->d_sb;
struct ocfs2_super *osb = OCFS2_SB(sb);
u32 tmp;
- if (!ocfs2_parse_options(sb, data, &parsed_options, 1) ||
- !ocfs2_check_set_options(sb, &parsed_options)) {
+ sync_filesystem(sb);
+
+ if (!ocfs2_check_set_options(sb, parsed_options)) {
ret = -EINVAL;
goto out;
}
tmp = OCFS2_MOUNT_HB_LOCAL | OCFS2_MOUNT_HB_GLOBAL |
OCFS2_MOUNT_HB_NONE;
- if ((osb->s_mount_opt & tmp) != (parsed_options.mount_opt & tmp)) {
+ if ((osb->s_mount_opt & tmp) != (parsed_options->mount_opt & tmp)) {
ret = -EINVAL;
mlog(ML_ERROR, "Cannot change heartbeat mode on remount\n");
goto out;
}
if ((osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK) !=
- (parsed_options.mount_opt & OCFS2_MOUNT_DATA_WRITEBACK)) {
+ (parsed_options->mount_opt & OCFS2_MOUNT_DATA_WRITEBACK)) {
ret = -EINVAL;
mlog(ML_ERROR, "Cannot change data mode on remount\n");
goto out;
@@ -655,16 +638,16 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
/* Probably don't want this on remount; it might
* mess with other nodes */
if (!(osb->s_mount_opt & OCFS2_MOUNT_INODE64) &&
- (parsed_options.mount_opt & OCFS2_MOUNT_INODE64)) {
+ (parsed_options->mount_opt & OCFS2_MOUNT_INODE64)) {
ret = -EINVAL;
mlog(ML_ERROR, "Cannot enable inode64 on remount\n");
goto out;
}
/* We're going to/from readonly mode. */
- if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY)) {
+ if ((bool)(fc->sb_flags & SB_RDONLY) != sb_rdonly(sb)) {
/* Disable quota accounting before remounting RO */
- if (*flags & MS_RDONLY) {
+ if (fc->sb_flags & SB_RDONLY) {
ret = ocfs2_susp_quotas(osb, 0);
if (ret < 0)
goto out;
@@ -678,8 +661,8 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
goto unlock_osb;
}
- if (*flags & MS_RDONLY) {
- sb->s_flags |= MS_RDONLY;
+ if (fc->sb_flags & SB_RDONLY) {
+ sb->s_flags |= SB_RDONLY;
osb->osb_flags |= OCFS2_OSB_SOFT_RO;
} else {
if (osb->osb_flags & OCFS2_OSB_ERROR_FS) {
@@ -696,14 +679,14 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
ret = -EINVAL;
goto unlock_osb;
}
- sb->s_flags &= ~MS_RDONLY;
+ sb->s_flags &= ~SB_RDONLY;
osb->osb_flags &= ~OCFS2_OSB_SOFT_RO;
}
- trace_ocfs2_remount(sb->s_flags, osb->osb_flags, *flags);
+ trace_ocfs2_remount(sb->s_flags, osb->osb_flags, fc->sb_flags);
unlock_osb:
spin_unlock(&osb->osb_lock);
/* Enable quota accounting after remounting RW */
- if (!ret && !(*flags & MS_RDONLY)) {
+ if (!ret && !(fc->sb_flags & SB_RDONLY)) {
if (sb_any_quota_suspended(sb))
ret = ocfs2_susp_quotas(osb, 1);
else
@@ -711,7 +694,7 @@ unlock_osb:
if (ret < 0) {
/* Return back changes... */
spin_lock(&osb->osb_lock);
- sb->s_flags |= MS_RDONLY;
+ sb->s_flags |= SB_RDONLY;
osb->osb_flags |= OCFS2_OSB_SOFT_RO;
spin_unlock(&osb->osb_lock);
goto out;
@@ -722,18 +705,18 @@ unlock_osb:
if (!ret) {
/* Only save off the new mount options in case of a successful
* remount. */
- osb->s_mount_opt = parsed_options.mount_opt;
- osb->s_atime_quantum = parsed_options.atime_quantum;
- osb->preferred_slot = parsed_options.slot;
- if (parsed_options.commit_interval)
- osb->osb_commit_interval = parsed_options.commit_interval;
+ osb->s_mount_opt = parsed_options->mount_opt;
+ osb->s_atime_quantum = parsed_options->atime_quantum;
+ osb->preferred_slot = parsed_options->slot;
+ if (parsed_options->commit_interval)
+ osb->osb_commit_interval = parsed_options->commit_interval;
if (!ocfs2_is_hard_readonly(osb))
ocfs2_set_journal_params(osb);
- sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+ sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ?
- MS_POSIXACL : 0);
+ SB_POSIXACL : 0);
}
out:
return ret;
@@ -894,11 +877,12 @@ static int ocfs2_susp_quotas(struct ocfs2_super *osb, int unsuspend)
{
int type;
struct super_block *sb = osb->sb;
- unsigned int feature[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA};
+ unsigned int feature[OCFS2_MAXQUOTAS] = {
+ OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
+ OCFS2_FEATURE_RO_COMPAT_GRPQUOTA};
int status = 0;
- for (type = 0; type < MAXQUOTAS; type++) {
+ for (type = 0; type < OCFS2_MAXQUOTAS; type++) {
if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
continue;
if (unsuspend)
@@ -922,17 +906,19 @@ static int ocfs2_susp_quotas(struct ocfs2_super *osb, int unsuspend)
static int ocfs2_enable_quotas(struct ocfs2_super *osb)
{
- struct inode *inode[MAXQUOTAS] = { NULL, NULL };
+ struct inode *inode[OCFS2_MAXQUOTAS] = { NULL, NULL };
struct super_block *sb = osb->sb;
- unsigned int feature[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA};
- unsigned int ino[MAXQUOTAS] = { LOCAL_USER_QUOTA_SYSTEM_INODE,
+ unsigned int feature[OCFS2_MAXQUOTAS] = {
+ OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
+ OCFS2_FEATURE_RO_COMPAT_GRPQUOTA};
+ unsigned int ino[OCFS2_MAXQUOTAS] = {
+ LOCAL_USER_QUOTA_SYSTEM_INODE,
LOCAL_GROUP_QUOTA_SYSTEM_INODE };
int status;
int type;
sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE | DQUOT_NEGATIVE_USAGE;
- for (type = 0; type < MAXQUOTAS; type++) {
+ for (type = 0; type < OCFS2_MAXQUOTAS; type++) {
if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
continue;
inode[type] = ocfs2_get_system_file_inode(osb, ino[type],
@@ -941,18 +927,18 @@ static int ocfs2_enable_quotas(struct ocfs2_super *osb)
status = -ENOENT;
goto out_quota_off;
}
- status = dquot_enable(inode[type], type, QFMT_OCFS2,
- DQUOT_USAGE_ENABLED);
+ status = dquot_load_quota_inode(inode[type], type, QFMT_OCFS2,
+ DQUOT_USAGE_ENABLED);
if (status < 0)
goto out_quota_off;
}
- for (type = 0; type < MAXQUOTAS; type++)
+ for (type = 0; type < OCFS2_MAXQUOTAS; type++)
iput(inode[type]);
return 0;
out_quota_off:
ocfs2_disable_quotas(osb);
- for (type = 0; type < MAXQUOTAS; type++)
+ for (type = 0; type < OCFS2_MAXQUOTAS; type++)
iput(inode[type]);
mlog_errno(status);
return status;
@@ -967,121 +953,85 @@ static void ocfs2_disable_quotas(struct ocfs2_super *osb)
/* We mostly ignore errors in this function because there's not much
* we can do when we see them */
- for (type = 0; type < MAXQUOTAS; type++) {
+ for (type = 0; type < OCFS2_MAXQUOTAS; type++) {
if (!sb_has_quota_loaded(sb, type))
continue;
- /* Cancel periodic syncing before we grab dqonoff_mutex */
- oinfo = sb_dqinfo(sb, type)->dqi_priv;
- cancel_delayed_work_sync(&oinfo->dqi_sync_work);
+ if (!sb_has_quota_suspended(sb, type)) {
+ oinfo = sb_dqinfo(sb, type)->dqi_priv;
+ cancel_delayed_work_sync(&oinfo->dqi_sync_work);
+ }
inode = igrab(sb->s_dquot.files[type]);
/* Turn off quotas. This will remove all dquot structures from
* memory and so they will be automatically synced to global
* quota files */
dquot_disable(sb, type, DQUOT_USAGE_ENABLED |
DQUOT_LIMITS_ENABLED);
- if (!inode)
- continue;
iput(inode);
}
}
-/* Handle quota on quotactl */
-static int ocfs2_quota_on(struct super_block *sb, int type, int format_id)
-{
- unsigned int feature[MAXQUOTAS] = { OCFS2_FEATURE_RO_COMPAT_USRQUOTA,
- OCFS2_FEATURE_RO_COMPAT_GRPQUOTA};
-
- if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
- return -EINVAL;
-
- return dquot_enable(sb_dqopt(sb)->files[type], type,
- format_id, DQUOT_LIMITS_ENABLED);
-}
-
-/* Handle quota off quotactl */
-static int ocfs2_quota_off(struct super_block *sb, int type)
-{
- return dquot_disable(sb, type, DQUOT_LIMITS_ENABLED);
-}
-
-static const struct quotactl_ops ocfs2_quotactl_ops = {
- .quota_on_meta = ocfs2_quota_on,
- .quota_off = ocfs2_quota_off,
- .quota_sync = dquot_quota_sync,
- .get_info = dquot_get_dqinfo,
- .set_info = dquot_set_dqinfo,
- .get_dqblk = dquot_get_dqblk,
- .set_dqblk = dquot_set_dqblk,
-};
-
-static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
+static int ocfs2_fill_super(struct super_block *sb, struct fs_context *fc)
{
struct dentry *root;
int status, sector_size;
- struct mount_options parsed_options;
+ struct mount_options *parsed_options = fc->fs_private;
struct inode *inode = NULL;
struct ocfs2_super *osb = NULL;
struct buffer_head *bh = NULL;
- char nodestr[8];
+ char nodestr[12];
struct ocfs2_blockcheck_stats stats;
- trace_ocfs2_fill_super(sb, data, silent);
-
- if (!ocfs2_parse_options(sb, data, &parsed_options, 0)) {
- status = -EINVAL;
- goto read_super_error;
- }
+ trace_ocfs2_fill_super(sb, fc, fc->sb_flags & SB_SILENT);
/* probe for superblock */
status = ocfs2_sb_probe(sb, &bh, &sector_size, &stats);
if (status < 0) {
mlog(ML_ERROR, "superblock probe failed!\n");
- goto read_super_error;
+ goto out;
}
status = ocfs2_initialize_super(sb, bh, sector_size, &stats);
- osb = OCFS2_SB(sb);
- if (status < 0) {
- mlog_errno(status);
- goto read_super_error;
- }
brelse(bh);
bh = NULL;
+ if (status < 0)
+ goto out;
+
+ osb = OCFS2_SB(sb);
- if (!ocfs2_check_set_options(sb, &parsed_options)) {
+ 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;
-
- ocfs2_la_set_sizes(osb, parsed_options.localalloc_opt);
- osb->osb_resv_level = parsed_options.resv_level;
- osb->osb_dir_resv_level = parsed_options.resv_level;
- if (parsed_options.dir_resv_level == -1)
- osb->osb_dir_resv_level = parsed_options.resv_level;
+ goto out_super;
+ }
+ 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;
+
+ ocfs2_la_set_sizes(osb, parsed_options->localalloc_opt);
+ osb->osb_resv_level = parsed_options->resv_level;
+ osb->osb_dir_resv_level = parsed_options->resv_level;
+ if (parsed_options->dir_resv_level == -1)
+ osb->osb_dir_resv_level = parsed_options->resv_level;
else
- osb->osb_dir_resv_level = parsed_options.dir_resv_level;
+ osb->osb_dir_resv_level = parsed_options->dir_resv_level;
- status = ocfs2_verify_userspace_stack(osb, &parsed_options);
+ status = ocfs2_verify_userspace_stack(osb, parsed_options);
if (status)
- goto read_super_error;
+ goto out_super;
sb->s_magic = OCFS2_SUPER_MAGIC;
- sb->s_flags = (sb->s_flags & ~(MS_POSIXACL | MS_NOSEC)) |
- ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
+ sb->s_flags = (sb->s_flags & ~(SB_POSIXACL | SB_NOSEC)) |
+ ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? SB_POSIXACL : 0);
- /* Hard readonly mode only if: bdev_read_only, MS_RDONLY,
+ /* Hard readonly mode only if: bdev_read_only, SB_RDONLY,
* heartbeat=none */
if (bdev_read_only(sb->s_bdev)) {
- if (!(sb->s_flags & MS_RDONLY)) {
+ if (!sb_rdonly(sb)) {
status = -EACCES;
mlog(ML_ERROR, "Readonly device detected but readonly "
"mount was not specified.\n");
- goto read_super_error;
+ goto out_super;
}
/* You should not be able to start a local heartbeat
@@ -1090,7 +1040,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
status = -EROFS;
mlog(ML_ERROR, "Local heartbeat specified on readonly "
"device.\n");
- goto read_super_error;
+ goto out_super;
}
status = ocfs2_check_journals_nolocks(osb);
@@ -1099,9 +1049,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
mlog(ML_ERROR, "Recovery required on readonly "
"file system, but write access is "
"unavailable.\n");
- else
- mlog_errno(status);
- goto read_super_error;
+ goto out_super;
}
ocfs2_set_ro_flag(osb, 1);
@@ -1112,64 +1060,59 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
}
if (!ocfs2_is_hard_readonly(osb)) {
- if (sb->s_flags & MS_RDONLY)
+ if (sb_rdonly(sb))
ocfs2_set_ro_flag(osb, 0);
}
status = ocfs2_verify_heartbeat(osb);
- if (status < 0) {
- mlog_errno(status);
- goto read_super_error;
- }
+ if (status < 0)
+ goto out_super;
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;
- }
- osb->osb_ctxt = debugfs_create_file("fs_state", S_IFREG|S_IRUSR,
- osb->osb_debug_root,
- osb,
- &ocfs2_osb_debug_fops);
- if (!osb->osb_ctxt) {
- status = -EINVAL;
- mlog_errno(status);
- goto read_super_error;
- }
+ debugfs_create_file("fs_state", S_IFREG|S_IRUSR, osb->osb_debug_root,
+ osb, &ocfs2_osb_debug_fops);
if (ocfs2_meta_ecc(osb)) {
- status = ocfs2_blockcheck_stats_debugfs_install(
- &osb->osb_ecc_stats,
- osb->osb_debug_root);
- if (status) {
- mlog(ML_ERROR,
- "Unable to create blockcheck statistics "
- "files\n");
- goto read_super_error;
- }
+ ocfs2_initialize_journal_triggers(sb, osb->s_journal_triggers);
+ ocfs2_blockcheck_stats_debugfs_install( &osb->osb_ecc_stats,
+ osb->osb_debug_root);
}
status = ocfs2_mount_volume(sb);
if (status < 0)
- goto read_super_error;
+ goto out_debugfs;
if (osb->root_inode)
inode = igrab(osb->root_inode);
if (!inode) {
status = -EIO;
- mlog_errno(status);
- goto read_super_error;
+ goto out_dismount;
+ }
+
+ osb->osb_dev_kset = kset_create_and_add(sb->s_id, NULL,
+ &ocfs2_kset->kobj);
+ if (!osb->osb_dev_kset) {
+ status = -ENOMEM;
+ mlog(ML_ERROR, "Unable to create device kset %s.\n", sb->s_id);
+ goto out_dismount;
+ }
+
+ /* Create filecheck sysfs related directories/files at
+ * /sys/fs/ocfs2/<devname>/filecheck */
+ if (ocfs2_filecheck_create_sysfs(osb)) {
+ status = -ENOMEM;
+ mlog(ML_ERROR, "Unable to create filecheck sysfs directory at "
+ "/sys/fs/ocfs2/%s/filecheck.\n", sb->s_id);
+ goto out_dismount;
}
root = d_make_root(inode);
if (!root) {
status = -ENOMEM;
- mlog_errno(status);
- goto read_super_error;
+ goto out_dismount;
}
sb->s_root = root;
@@ -1193,7 +1136,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
/* Now we can initialize quotas because we can afford to wait
* for cluster locks recovery now. That also means that truncation
* log recovery can happen but that waits for proper quota setup */
- if (!(sb->s_flags & MS_RDONLY)) {
+ if (!sb_rdonly(sb)) {
status = ocfs2_enable_quotas(osb);
if (status < 0) {
/* We have to err-out specially here because
@@ -1216,60 +1159,92 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
return status;
-read_super_error:
- brelse(bh);
+out_dismount:
+ atomic_set(&osb->vol_state, VOLUME_DISABLED);
+ wake_up(&osb->osb_mount_event);
+ ocfs2_free_replay_slots(osb);
+ ocfs2_dismount_volume(sb, 1);
+ goto out;
- if (osb) {
- atomic_set(&osb->vol_state, VOLUME_DISABLED);
- wake_up(&osb->osb_mount_event);
- ocfs2_dismount_volume(sb, 1);
- }
+out_debugfs:
+ debugfs_remove_recursive(osb->osb_debug_root);
+out_super:
+ ocfs2_release_system_inodes(osb);
+ kfree(osb->recovery_map);
+ ocfs2_delete_osb(osb);
+ kfree(osb);
+out:
+ mlog_errno(status);
- if (status)
- mlog_errno(status);
return status;
}
-static struct dentry *ocfs2_mount(struct file_system_type *fs_type,
- int flags,
- const char *dev_name,
- void *data)
+static int ocfs2_get_tree(struct fs_context *fc)
{
- return mount_bdev(fs_type, flags, dev_name, data, ocfs2_fill_super);
+ return get_tree_bdev(fc, ocfs2_fill_super);
}
-static void ocfs2_kill_sb(struct super_block *sb)
+static void ocfs2_free_fc(struct fs_context *fc)
{
- struct ocfs2_super *osb = OCFS2_SB(sb);
+ kfree(fc->fs_private);
+}
- /* Failed mount? */
- if (!osb || atomic_read(&osb->vol_state) == VOLUME_DISABLED)
- goto out;
+static const struct fs_context_operations ocfs2_context_ops = {
+ .parse_param = ocfs2_parse_param,
+ .get_tree = ocfs2_get_tree,
+ .reconfigure = ocfs2_reconfigure,
+ .free = ocfs2_free_fc,
+};
- /* Prevent further queueing of inode drop events */
- spin_lock(&dentry_list_lock);
- ocfs2_set_osb_flag(osb, OCFS2_OSB_DROP_DENTRY_LOCK_IMMED);
- spin_unlock(&dentry_list_lock);
- /* Wait for work to finish and/or remove it */
- cancel_work_sync(&osb->dentry_lock_work);
-out:
- kill_block_super(sb);
+static int ocfs2_init_fs_context(struct fs_context *fc)
+{
+ struct mount_options *mopt;
+
+ mopt = kzalloc(sizeof(struct mount_options), GFP_KERNEL);
+ if (!mopt)
+ return -EINVAL;
+
+ mopt->commit_interval = 0;
+ mopt->mount_opt = OCFS2_MOUNT_NOINTR;
+ mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
+ mopt->slot = OCFS2_INVALID_SLOT;
+ mopt->localalloc_opt = -1;
+ mopt->cluster_stack[0] = '\0';
+ mopt->resv_level = OCFS2_DEFAULT_RESV_LEVEL;
+ mopt->dir_resv_level = -1;
+
+ fc->fs_private = mopt;
+ fc->ops = &ocfs2_context_ops;
+
+ return 0;
}
static struct file_system_type ocfs2_fs_type = {
.owner = THIS_MODULE,
.name = "ocfs2",
- .mount = ocfs2_mount,
- .kill_sb = ocfs2_kill_sb,
-
+ .kill_sb = kill_block_super,
.fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE,
- .next = NULL
+ .next = NULL,
+ .init_fs_context = ocfs2_init_fs_context,
+ .parameters = ocfs2_param_spec,
};
MODULE_ALIAS_FS("ocfs2");
static int ocfs2_check_set_options(struct super_block *sb,
struct mount_options *options)
{
+ if (options->user_stack == 0) {
+ u32 tmp;
+
+ /* Ensure only one heartbeat mode */
+ tmp = options->mount_opt & (OCFS2_MOUNT_HB_LOCAL |
+ OCFS2_MOUNT_HB_GLOBAL |
+ OCFS2_MOUNT_HB_NONE);
+ if (hweight32(tmp) != 1) {
+ mlog(ML_ERROR, "Invalid heartbeat mount options\n");
+ return 0;
+ }
+ }
if (options->mount_opt & OCFS2_MOUNT_USRQUOTA &&
!OCFS2_HAS_RO_COMPAT_FEATURE(sb,
OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
@@ -1301,233 +1276,142 @@ static int ocfs2_check_set_options(struct super_block *sb,
return 1;
}
-static int ocfs2_parse_options(struct super_block *sb,
- char *options,
- struct mount_options *mopt,
- int is_remount)
+static int ocfs2_parse_param(struct fs_context *fc, struct fs_parameter *param)
{
- int status, user_stack = 0;
- char *p;
- u32 tmp;
-
- trace_ocfs2_parse_options(is_remount, options ? options : "(none)");
-
- mopt->commit_interval = 0;
- mopt->mount_opt = OCFS2_MOUNT_NOINTR;
- mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
- mopt->slot = OCFS2_INVALID_SLOT;
- mopt->localalloc_opt = -1;
- mopt->cluster_stack[0] = '\0';
- mopt->resv_level = OCFS2_DEFAULT_RESV_LEVEL;
- mopt->dir_resv_level = -1;
-
- 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:
- mopt->mount_opt |= OCFS2_MOUNT_HB_LOCAL;
- break;
- case Opt_hb_none:
- mopt->mount_opt |= OCFS2_MOUNT_HB_NONE;
- break;
- case Opt_hb_global:
- mopt->mount_opt |= OCFS2_MOUNT_HB_GLOBAL;
- break;
- case Opt_barrier:
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option)
- mopt->mount_opt |= OCFS2_MOUNT_BARRIER;
- else
- mopt->mount_opt &= ~OCFS2_MOUNT_BARRIER;
- break;
- case Opt_intr:
- mopt->mount_opt &= ~OCFS2_MOUNT_NOINTR;
- break;
- case Opt_nointr:
+ struct fs_parse_result result;
+ int opt;
+ struct mount_options *mopt = fc->fs_private;
+ bool is_remount = (fc->purpose & FS_CONTEXT_FOR_RECONFIGURE);
+
+ trace_ocfs2_parse_options(is_remount, param->key);
+
+ opt = fs_parse(fc, ocfs2_param_spec, param, &result);
+ if (opt < 0)
+ return opt;
+
+ switch (opt) {
+ case Opt_heartbeat:
+ mopt->mount_opt |= result.uint_32;
+ break;
+ case Opt_barrier:
+ if (result.uint_32)
+ mopt->mount_opt |= OCFS2_MOUNT_BARRIER;
+ else
+ mopt->mount_opt &= ~OCFS2_MOUNT_BARRIER;
+ break;
+ case Opt_intr:
+ if (result.negated)
mopt->mount_opt |= OCFS2_MOUNT_NOINTR;
- break;
- case Opt_err_panic:
- mopt->mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
- break;
- case Opt_err_ro:
- mopt->mount_opt &= ~OCFS2_MOUNT_ERRORS_PANIC;
- break;
- case Opt_data_ordered:
- mopt->mount_opt &= ~OCFS2_MOUNT_DATA_WRITEBACK;
- break;
- case Opt_data_writeback:
- mopt->mount_opt |= OCFS2_MOUNT_DATA_WRITEBACK;
- break;
- case Opt_user_xattr:
- mopt->mount_opt &= ~OCFS2_MOUNT_NOUSERXATTR;
- break;
- case Opt_nouser_xattr:
+ else
+ mopt->mount_opt &= ~OCFS2_MOUNT_NOINTR;
+ break;
+ case Opt_errors:
+ mopt->mount_opt &= ~(OCFS2_MOUNT_ERRORS_CONT |
+ OCFS2_MOUNT_ERRORS_ROFS |
+ OCFS2_MOUNT_ERRORS_PANIC);
+ mopt->mount_opt |= result.uint_32;
+ break;
+ case Opt_data:
+ mopt->mount_opt &= ~OCFS2_MOUNT_DATA_WRITEBACK;
+ mopt->mount_opt |= result.uint_32;
+ break;
+ case Opt_user_xattr:
+ if (result.negated)
mopt->mount_opt |= OCFS2_MOUNT_NOUSERXATTR;
- break;
- case Opt_atime_quantum:
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option >= 0)
- mopt->atime_quantum = option;
- break;
- case Opt_slot:
- option = 0;
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option)
- mopt->slot = (s16)option;
- break;
- case Opt_commit:
- option = 0;
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option < 0)
- return 0;
- if (option == 0)
- option = JBD2_DEFAULT_MAX_COMMIT_AGE;
- mopt->commit_interval = HZ * option;
- break;
- case Opt_localalloc:
- option = 0;
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option >= 0)
- mopt->localalloc_opt = option;
- break;
- 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;
- case Opt_stack:
- /* Check both that the option we were passed
- * is of the right length and that it is a proper
- * string of the right length.
- */
- if (((args[0].to - args[0].from) !=
- OCFS2_STACK_LABEL_LEN) ||
- (strnlen(args[0].from,
- OCFS2_STACK_LABEL_LEN) !=
- OCFS2_STACK_LABEL_LEN)) {
- mlog(ML_ERROR,
- "Invalid cluster_stack option\n");
- status = 0;
- goto bail;
- }
- memcpy(mopt->cluster_stack, args[0].from,
- OCFS2_STACK_LABEL_LEN);
- mopt->cluster_stack[OCFS2_STACK_LABEL_LEN] = '\0';
- /*
- * Open code the memcmp here as we don't have
- * an osb to pass to
- * ocfs2_userspace_stack().
- */
- if (memcmp(mopt->cluster_stack,
- OCFS2_CLASSIC_CLUSTER_STACK,
- OCFS2_STACK_LABEL_LEN))
- user_stack = 1;
- break;
- case Opt_inode64:
- mopt->mount_opt |= OCFS2_MOUNT_INODE64;
- break;
- case Opt_usrquota:
- mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA;
- break;
- case Opt_grpquota:
- mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA;
- break;
- case Opt_coherency_buffered:
- mopt->mount_opt |= OCFS2_MOUNT_COHERENCY_BUFFERED;
- break;
- case Opt_coherency_full:
- mopt->mount_opt &= ~OCFS2_MOUNT_COHERENCY_BUFFERED;
- break;
- case Opt_acl:
- mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
- mopt->mount_opt &= ~OCFS2_MOUNT_NO_POSIX_ACL;
- break;
- case Opt_noacl:
+ else
+ mopt->mount_opt &= ~OCFS2_MOUNT_NOUSERXATTR;
+ break;
+ case Opt_atime_quantum:
+ mopt->atime_quantum = result.uint_32;
+ break;
+ case Opt_slot:
+ if (result.uint_32)
+ mopt->slot = (u16)result.uint_32;
+ break;
+ case Opt_commit:
+ if (result.uint_32 == 0)
+ mopt->commit_interval = HZ * JBD2_DEFAULT_MAX_COMMIT_AGE;
+ else
+ mopt->commit_interval = HZ * result.uint_32;
+ break;
+ case Opt_localalloc:
+ if (result.int_32 >= 0)
+ mopt->localalloc_opt = result.int_32;
+ break;
+ 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;
+ case Opt_stack:
+ /* Check both that the option we were passed is of the right
+ * length and that it is a proper string of the right length.
+ */
+ if (strlen(param->string) != OCFS2_STACK_LABEL_LEN) {
+ mlog(ML_ERROR, "Invalid cluster_stack option\n");
+ return -EINVAL;
+ }
+ memcpy(mopt->cluster_stack, param->string, OCFS2_STACK_LABEL_LEN);
+ mopt->cluster_stack[OCFS2_STACK_LABEL_LEN] = '\0';
+ /*
+ * Open code the memcmp here as we don't have an osb to pass
+ * to ocfs2_userspace_stack().
+ */
+ if (memcmp(mopt->cluster_stack,
+ OCFS2_CLASSIC_CLUSTER_STACK,
+ OCFS2_STACK_LABEL_LEN))
+ mopt->user_stack = 1;
+ break;
+ case Opt_inode64:
+ mopt->mount_opt |= OCFS2_MOUNT_INODE64;
+ break;
+ case Opt_usrquota:
+ mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA;
+ break;
+ case Opt_grpquota:
+ mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA;
+ break;
+ case Opt_coherency:
+ mopt->mount_opt &= ~OCFS2_MOUNT_COHERENCY_BUFFERED;
+ mopt->mount_opt |= result.uint_32;
+ break;
+ case Opt_acl:
+ if (result.negated) {
mopt->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL;
mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
+ } else {
+ mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+ mopt->mount_opt &= ~OCFS2_MOUNT_NO_POSIX_ACL;
+ }
+ break;
+ case Opt_resv_level:
+ if (is_remount)
break;
- case Opt_resv_level:
- if (is_remount)
- break;
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option >= OCFS2_MIN_RESV_LEVEL &&
- option < OCFS2_MAX_RESV_LEVEL)
- mopt->resv_level = option;
- break;
- case Opt_dir_resv_level:
- if (is_remount)
- break;
- if (match_int(&args[0], &option)) {
- status = 0;
- goto bail;
- }
- if (option >= OCFS2_MIN_RESV_LEVEL &&
- option < OCFS2_MAX_RESV_LEVEL)
- mopt->dir_resv_level = option;
+ if (result.uint_32 >= OCFS2_MIN_RESV_LEVEL &&
+ result.uint_32 < OCFS2_MAX_RESV_LEVEL)
+ mopt->resv_level = result.uint_32;
+ break;
+ case Opt_dir_resv_level:
+ if (is_remount)
break;
- default:
- mlog(ML_ERROR,
- "Unrecognized mount option \"%s\" "
- "or missing value\n", p);
- status = 0;
- goto bail;
- }
- }
-
- if (user_stack == 0) {
- /* Ensure only one heartbeat mode */
- tmp = mopt->mount_opt & (OCFS2_MOUNT_HB_LOCAL |
- OCFS2_MOUNT_HB_GLOBAL |
- OCFS2_MOUNT_HB_NONE);
- if (hweight32(tmp) != 1) {
- mlog(ML_ERROR, "Invalid heartbeat mount options\n");
- status = 0;
- goto bail;
- }
+ if (result.uint_32 >= OCFS2_MIN_RESV_LEVEL &&
+ result.uint_32 < OCFS2_MAX_RESV_LEVEL)
+ mopt->dir_resv_level = result.uint_32;
+ break;
+ case Opt_journal_async_commit:
+ mopt->mount_opt |= OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT;
+ break;
+ default:
+ return -EINVAL;
}
- status = 1;
-
-bail:
- return status;
+ return 0;
}
static int ocfs2_show_options(struct seq_file *s, struct dentry *root)
@@ -1558,6 +1442,8 @@ static int ocfs2_show_options(struct seq_file *s, struct dentry *root)
if (opts & OCFS2_MOUNT_ERRORS_PANIC)
seq_printf(s, ",errors=panic");
+ else if (opts & OCFS2_MOUNT_ERRORS_CONT)
+ seq_printf(s, ",errors=continue");
else
seq_printf(s, ",errors=remount-ro");
@@ -1578,8 +1464,7 @@ static int ocfs2_show_options(struct seq_file *s, struct dentry *root)
seq_printf(s, ",localflocks,");
if (osb->osb_cluster_stack[0])
- seq_printf(s, ",cluster_stack=%.*s", OCFS2_STACK_LABEL_LEN,
- osb->osb_cluster_stack);
+ seq_show_option(s, "cluster_stack", osb->osb_cluster_stack);
if (opts & OCFS2_MOUNT_USRQUOTA)
seq_printf(s, ",usrquota");
if (opts & OCFS2_MOUNT_GRPQUOTA)
@@ -1609,19 +1494,15 @@ static int ocfs2_show_options(struct seq_file *s, struct dentry *root)
if (osb->osb_dir_resv_level != osb->osb_resv_level)
seq_printf(s, ",dir_resv_level=%d", osb->osb_resv_level);
+ if (opts & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT)
+ seq_printf(s, ",journal_async_commit");
+
return 0;
}
-wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ];
-
static int __init ocfs2_init(void)
{
- int status, i;
-
- ocfs2_print_version();
-
- for (i = 0; i < OCFS2_IOEND_WQ_HASH_SZ; i++)
- init_waitqueue_head(&ocfs2__ioend_wq[i]);
+ int status;
status = init_ocfs2_uptodate_cache();
if (status < 0)
@@ -1631,32 +1512,18 @@ static int __init ocfs2_init(void)
if (status < 0)
goto out2;
- ocfs2_wq = create_singlethread_workqueue("ocfs2_wq");
- if (!ocfs2_wq) {
- status = -ENOMEM;
- goto out3;
- }
-
ocfs2_debugfs_root = debugfs_create_dir("ocfs2", NULL);
- if (!ocfs2_debugfs_root) {
- status = -EFAULT;
- mlog(ML_ERROR, "Unable to create ocfs2 debugfs root.\n");
- }
ocfs2_set_locking_protocol();
- status = register_quota_format(&ocfs2_quota_format);
- if (status < 0)
- goto out4;
+ register_quota_format(&ocfs2_quota_format);
+
status = register_filesystem(&ocfs2_fs_type);
if (!status)
return 0;
unregister_quota_format(&ocfs2_quota_format);
-out4:
- destroy_workqueue(ocfs2_wq);
debugfs_remove(ocfs2_debugfs_root);
-out3:
ocfs2_free_mem_caches();
out2:
exit_ocfs2_uptodate_cache();
@@ -1667,11 +1534,6 @@ out1:
static void __exit ocfs2_exit(void)
{
- if (ocfs2_wq) {
- flush_workqueue(ocfs2_wq);
- destroy_workqueue(ocfs2_wq);
- }
-
unregister_quota_format(&ocfs2_quota_format);
debugfs_remove(ocfs2_debugfs_root);
@@ -1744,8 +1606,7 @@ static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
ocfs2_inode_unlock(inode, 0);
status = 0;
bail:
- if (inode)
- iput(inode);
+ iput(inode);
if (status)
mlog_errno(status);
@@ -1762,14 +1623,15 @@ static void ocfs2_inode_init_once(void *data)
spin_lock_init(&oi->ip_lock);
ocfs2_extent_map_init(&oi->vfs_inode);
INIT_LIST_HEAD(&oi->ip_io_markers);
+ INIT_LIST_HEAD(&oi->ip_unwritten_list);
oi->ip_dir_start_lookup = 0;
- atomic_set(&oi->ip_unaligned_aio, 0);
init_rwsem(&oi->ip_alloc_sem);
init_rwsem(&oi->ip_xattr_sem);
mutex_init(&oi->ip_io_mutex);
oi->ip_blkno = 0ULL;
oi->ip_clusters = 0;
+ oi->ip_next_orphan = NULL;
ocfs2_resv_init_once(&oi->ip_la_data_resv);
@@ -1789,27 +1651,23 @@ static int ocfs2_initialize_mem_caches(void)
sizeof(struct ocfs2_inode_info),
0,
(SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
- SLAB_MEM_SPREAD),
+ SLAB_ACCOUNT),
ocfs2_inode_init_once);
ocfs2_dquot_cachep = kmem_cache_create("ocfs2_dquot_cache",
sizeof(struct ocfs2_dquot),
0,
- (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
- SLAB_MEM_SPREAD),
+ SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT,
NULL);
ocfs2_qf_chunk_cachep = kmem_cache_create("ocfs2_qf_chunk_cache",
sizeof(struct ocfs2_quota_chunk),
0,
- (SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD),
+ SLAB_RECLAIM_ACCOUNT,
NULL);
if (!ocfs2_inode_cachep || !ocfs2_dquot_cachep ||
!ocfs2_qf_chunk_cachep) {
- if (ocfs2_inode_cachep)
- kmem_cache_destroy(ocfs2_inode_cachep);
- if (ocfs2_dquot_cachep)
- kmem_cache_destroy(ocfs2_dquot_cachep);
- if (ocfs2_qf_chunk_cachep)
- kmem_cache_destroy(ocfs2_qf_chunk_cachep);
+ kmem_cache_destroy(ocfs2_inode_cachep);
+ kmem_cache_destroy(ocfs2_dquot_cachep);
+ kmem_cache_destroy(ocfs2_qf_chunk_cachep);
return -ENOMEM;
}
@@ -1823,16 +1681,13 @@ static void ocfs2_free_mem_caches(void)
* destroy cache.
*/
rcu_barrier();
- if (ocfs2_inode_cachep)
- kmem_cache_destroy(ocfs2_inode_cachep);
+ kmem_cache_destroy(ocfs2_inode_cachep);
ocfs2_inode_cachep = NULL;
- if (ocfs2_dquot_cachep)
- kmem_cache_destroy(ocfs2_dquot_cachep);
+ kmem_cache_destroy(ocfs2_dquot_cachep);
ocfs2_dquot_cachep = NULL;
- if (ocfs2_qf_chunk_cachep)
- kmem_cache_destroy(ocfs2_qf_chunk_cachep);
+ kmem_cache_destroy(ocfs2_qf_chunk_cachep);
ocfs2_qf_chunk_cachep = NULL;
}
@@ -1848,16 +1703,14 @@ static int ocfs2_get_sector(struct super_block *sb,
*bh = sb_getblk(sb, block);
if (!*bh) {
- mlog_errno(-EIO);
- return -EIO;
+ mlog_errno(-ENOMEM);
+ return -ENOMEM;
}
lock_buffer(*bh);
if (!buffer_dirty(*bh))
clear_buffer_uptodate(*bh);
unlock_buffer(*bh);
- ll_rw_block(READ, 1, bh);
- wait_on_buffer(*bh);
- if (!buffer_uptodate(*bh)) {
+ if (bh_read(*bh, 0) < 0) {
mlog_errno(-EIO);
brelse(*bh);
*bh = NULL;
@@ -1870,53 +1723,71 @@ static int ocfs2_get_sector(struct super_block *sb,
static int ocfs2_mount_volume(struct super_block *sb)
{
int status = 0;
- int unlock_super = 0;
struct ocfs2_super *osb = OCFS2_SB(sb);
if (ocfs2_is_hard_readonly(osb))
- goto leave;
+ goto out;
+
+ mutex_init(&osb->obs_trim_fs_mutex);
status = ocfs2_dlm_init(osb);
if (status < 0) {
mlog_errno(status);
- goto leave;
+ if (status == -EBADR && ocfs2_userspace_stack(osb))
+ mlog(ML_ERROR, "couldn't mount because cluster name on"
+ " disk does not match the running cluster name.\n");
+ goto out;
}
status = ocfs2_super_lock(osb, 1);
if (status < 0) {
mlog_errno(status);
- goto leave;
+ goto out_dlm;
}
- 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;
+ goto out_super_lock;
}
/* load all node-local system inodes */
status = ocfs2_init_local_system_inodes(osb);
if (status < 0) {
mlog_errno(status);
- goto leave;
+ goto out_super_lock;
}
status = ocfs2_check_volume(osb);
if (status < 0) {
mlog_errno(status);
- goto leave;
+ goto out_system_inodes;
}
status = ocfs2_truncate_log_init(osb);
- if (status < 0)
+ if (status < 0) {
mlog_errno(status);
+ goto out_check_volume;
+ }
-leave:
- if (unlock_super)
- ocfs2_super_unlock(osb, 1);
+ ocfs2_super_unlock(osb, 1);
+ return 0;
+out_check_volume:
+ ocfs2_free_replay_slots(osb);
+out_system_inodes:
+ if (osb->local_alloc_state == OCFS2_LA_ENABLED)
+ ocfs2_shutdown_local_alloc(osb);
+ ocfs2_release_system_inodes(osb);
+ /* before journal shutdown, we should release slot_info */
+ ocfs2_free_slot_info(osb);
+ ocfs2_journal_shutdown(osb);
+out_super_lock:
+ ocfs2_super_unlock(osb, 1);
+out_dlm:
+ ocfs2_dlm_shutdown(osb, 0);
+out:
return status;
}
@@ -1924,7 +1795,7 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
{
int tmp, hangup_needed = 0;
struct ocfs2_super *osb = NULL;
- char nodestr[8];
+ char nodestr[12];
trace_ocfs2_dismount_volume(sb);
@@ -1932,19 +1803,25 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
osb = OCFS2_SB(sb);
BUG_ON(!osb);
- debugfs_remove(osb->osb_ctxt);
+ /* Remove file check sysfs related directories/files,
+ * and wait for the pending file check operations */
+ ocfs2_filecheck_remove_sysfs(osb);
- /*
- * Flush inode dropping work queue so that deletes are
- * performed while the filesystem is still working
- */
- ocfs2_drop_all_dl_inodes(osb);
+ kset_unregister(osb->osb_dev_kset);
/* Orphan scan should be stopped as early as possible */
ocfs2_orphan_scan_stop(osb);
+ /* Stop quota recovery so that we can disable quotas */
+ ocfs2_recovery_disable_quota(osb);
+
ocfs2_disable_quotas(osb);
+ /* All dquots should be freed by now */
+ WARN_ON(!llist_empty(&osb->dquot_drop_list));
+ /* Wait for worker to be done with the work structure in osb */
+ cancel_work_sync(&osb->dquot_drop_work);
+
ocfs2_shutdown_local_alloc(osb);
ocfs2_truncate_log_shutdown(osb);
@@ -1952,8 +1829,6 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
/* This will disable recovery and flush any recovery work. */
ocfs2_recovery_exit(osb);
- ocfs2_journal_shutdown(osb);
-
ocfs2_sync_blockdev(sb);
ocfs2_purge_refcount_trees(osb);
@@ -1976,6 +1851,8 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
ocfs2_release_system_inodes(osb);
+ ocfs2_journal_shutdown(osb);
+
/*
* If we're dismounting due to mount error, mount.ocfs2 will clean
* up heartbeat. If we're a local mount, there is no heartbeat.
@@ -1986,11 +1863,10 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
!ocfs2_is_hard_readonly(osb))
hangup_needed = 1;
- if (osb->cconn)
- ocfs2_dlm_shutdown(osb, hangup_needed);
+ ocfs2_dlm_shutdown(osb, hangup_needed);
ocfs2_blockcheck_stats_debugfs_remove(&osb->osb_ecc_stats);
- debugfs_remove(osb->osb_debug_root);
+ debugfs_remove_recursive(osb->osb_debug_root);
if (hangup_needed)
ocfs2_cluster_hangup(osb->uuid_str, strlen(osb->uuid_str));
@@ -2074,8 +1950,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
int i, cbits, bbits;
struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
struct inode *inode = NULL;
- struct ocfs2_journal *journal;
- __le32 uuid_net_key;
struct ocfs2_super *osb;
u64 total_blocks;
@@ -2083,22 +1957,25 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (!osb) {
status = -ENOMEM;
mlog_errno(status);
- goto bail;
+ goto out;
}
sb->s_fs_info = osb;
sb->s_op = &ocfs2_sops;
- sb->s_d_op = &ocfs2_dentry_ops;
+ set_default_d_op(sb, &ocfs2_dentry_ops);
sb->s_export_op = &ocfs2_export_ops;
- sb->s_qcop = &ocfs2_quotactl_ops;
+ sb->s_qcop = &dquot_quotactl_sysfile_ops;
sb->dq_op = &ocfs2_quota_operations;
+ sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
sb->s_xattr = ocfs2_xattr_handlers;
sb->s_time_gran = 1;
- sb->s_flags |= MS_NOATIME;
+ sb->s_flags |= SB_NOATIME;
/* this is needed to support O_LARGEFILE */
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);
+ super_set_uuid(sb, di->id2.i_super.s_uuid,
+ sizeof(di->id2.i_super.s_uuid));
osb->osb_dx_mask = (1 << (cbits - bbits)) - 1;
@@ -2107,7 +1984,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
osb->osb_dx_seed[3] = le32_to_cpu(di->id2.i_super.s_uuid_hash);
osb->sb = sb;
- /* Save off for ocfs2_rw_direct */
osb->s_sectsize_bits = blksize_bits(sector_size);
BUG_ON(!osb->s_sectsize_bits);
@@ -2121,6 +1997,8 @@ static int ocfs2_initialize_super(struct super_block *sb,
spin_lock_init(&osb->osb_xattr_lock);
ocfs2_init_steal_slots(osb);
+ mutex_init(&osb->system_file_mutex);
+
atomic_set(&osb->alloc_stats.moves, 0);
atomic_set(&osb->alloc_stats.local_data, 0);
atomic_set(&osb->alloc_stats.bitmap_data, 0);
@@ -2140,7 +2018,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
mlog(ML_ERROR, "Invalid number of node slots (%u)\n",
osb->max_slots);
status = -EINVAL;
- goto bail;
+ goto out;
}
ocfs2_orphan_scan_init(osb);
@@ -2149,7 +2027,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (status) {
mlog(ML_ERROR, "Unable to initialize recovery state\n");
mlog_errno(status);
- goto bail;
+ goto out;
}
init_waitqueue_head(&osb->checkpoint_event);
@@ -2167,17 +2045,13 @@ static int ocfs2_initialize_super(struct super_block *sb,
init_waitqueue_head(&osb->osb_mount_event);
- status = ocfs2_resmap_init(osb, &osb->osb_la_resmap);
- if (status) {
- mlog_errno(status);
- goto bail;
- }
+ ocfs2_resmap_init(osb, &osb->osb_la_resmap);
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;
+ goto out_recovery_map;
}
osb->slot_recovery_generations =
@@ -2186,7 +2060,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (!osb->slot_recovery_generations) {
status = -ENOMEM;
mlog_errno(status);
- goto bail;
+ goto out_vol_label;
}
init_waitqueue_head(&osb->osb_wipe_event);
@@ -2196,7 +2070,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (!osb->osb_orphan_wipes) {
status = -ENOMEM;
mlog_errno(status);
- goto bail;
+ goto out_slot_recovery_gen;
}
osb->osb_rf_lock_tree = RB_ROOT;
@@ -2212,31 +2086,38 @@ static int ocfs2_initialize_super(struct super_block *sb,
mlog(ML_ERROR, "couldn't mount because of unsupported "
"optional features (%x).\n", i);
status = -EINVAL;
- goto bail;
+ goto out_orphan_wipes;
}
- if (!(osb->sb->s_flags & MS_RDONLY) &&
- (i = OCFS2_HAS_RO_COMPAT_FEATURE(osb->sb, ~OCFS2_FEATURE_RO_COMPAT_SUPP))) {
+ if (!sb_rdonly(osb->sb) && (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;
+ goto out_orphan_wipes;
}
if (ocfs2_clusterinfo_valid(osb)) {
+ /*
+ * ci_stack and ci_cluster in ocfs2_cluster_info may not be null
+ * terminated, so make sure no overflow happens here by using
+ * memcpy. Destination strings will always be null terminated
+ * because osb is allocated using kzalloc.
+ */
osb->osb_stackflags =
OCFS2_RAW_SB(di)->s_cluster_info.ci_stackflags;
memcpy(osb->osb_cluster_stack,
OCFS2_RAW_SB(di)->s_cluster_info.ci_stack,
OCFS2_STACK_LABEL_LEN);
- osb->osb_cluster_stack[OCFS2_STACK_LABEL_LEN] = '\0';
if (strlen(osb->osb_cluster_stack) != OCFS2_STACK_LABEL_LEN) {
mlog(ML_ERROR,
"couldn't mount because of an invalid "
"cluster stack label (%s) \n",
osb->osb_cluster_stack);
status = -EINVAL;
- goto bail;
+ goto out_orphan_wipes;
}
+ memcpy(osb->osb_cluster_name,
+ OCFS2_RAW_SB(di)->s_cluster_info.ci_cluster,
+ OCFS2_CLUSTER_NAME_LEN);
} else {
/* The empty string is identical with classic tools that
* don't know about s_cluster_info. */
@@ -2245,35 +2126,17 @@ static int ocfs2_initialize_super(struct super_block *sb,
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.
+ /*
+ * FIXME
+ * This should be done in ocfs2_journal_init(), but any inode
+ * writes back operation will cause the filesystem to crash.
*/
- /* initialize our journal structure */
-
- journal = kzalloc(sizeof(struct ocfs2_journal), GFP_KERNEL);
- 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);
- INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery);
- journal->j_state = OCFS2_JOURNAL_FREE;
+ status = ocfs2_journal_alloc(osb);
+ if (status < 0)
+ goto out_orphan_wipes;
- INIT_WORK(&osb->dentry_lock_work, ocfs2_drop_dl_inodes);
- osb->dentry_lock_list = NULL;
+ INIT_WORK(&osb->dquot_drop_work, ocfs2_drop_dquot_refs);
+ init_llist_head(&osb->dquot_drop_list);
/* get some pseudo constants for clustersize bits */
osb->s_clustersize_bits =
@@ -2285,7 +2148,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
mlog(ML_ERROR, "Volume has invalid cluster size (%d)\n",
osb->s_clustersize);
status = -EINVAL;
- goto bail;
+ goto out_journal;
}
total_blocks = ocfs2_clusters_to_blocks(osb->sb,
@@ -2297,20 +2160,18 @@ static int ocfs2_initialize_super(struct super_block *sb,
mlog(ML_ERROR, "Volume too large "
"to mount safely on this system");
status = -EFBIG;
- goto bail;
+ goto out_journal;
}
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;
+ goto out_journal;
}
- memcpy(&uuid_net_key, di->id2.i_super.s_uuid, sizeof(uuid_net_key));
-
- strncpy(osb->vol_label, di->id2.i_super.s_label, 63);
- osb->vol_label[63] = '\0';
+ strscpy(osb->vol_label, di->id2.i_super.s_label,
+ OCFS2_MAX_VOL_LABEL_LEN);
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 =
@@ -2326,7 +2187,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (!osb->osb_dlm_debug) {
status = -ENOMEM;
mlog_errno(status);
- goto bail;
+ goto out_uuid_str;
}
atomic_set(&osb->vol_state, VOLUME_INIT);
@@ -2335,7 +2196,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
status = ocfs2_init_global_system_inodes(osb);
if (status < 0) {
mlog_errno(status);
- goto bail;
+ goto out_dlm_out;
}
/*
@@ -2346,7 +2207,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
if (!inode) {
status = -EINVAL;
mlog_errno(status);
- goto bail;
+ goto out_system_inodes;
}
osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno;
@@ -2359,11 +2220,39 @@ static int ocfs2_initialize_super(struct super_block *sb,
status = ocfs2_init_slot_info(osb);
if (status < 0) {
mlog_errno(status);
- goto bail;
+ goto out_system_inodes;
}
- cleancache_init_shared_fs((char *)&di->id2.i_super.s_uuid, sb);
-bail:
+ osb->ocfs2_wq = alloc_ordered_workqueue("ocfs2_wq", WQ_MEM_RECLAIM);
+ if (!osb->ocfs2_wq) {
+ status = -ENOMEM;
+ mlog_errno(status);
+ goto out_slot_info;
+ }
+
+ return status;
+
+out_slot_info:
+ ocfs2_free_slot_info(osb);
+out_system_inodes:
+ ocfs2_release_system_inodes(osb);
+out_dlm_out:
+ ocfs2_put_dlm_debug(osb->osb_dlm_debug);
+out_uuid_str:
+ kfree(osb->uuid_str);
+out_journal:
+ kfree(osb->journal);
+out_orphan_wipes:
+ kfree(osb->osb_orphan_wipes);
+out_slot_recovery_gen:
+ kfree(osb->slot_recovery_generations);
+out_vol_label:
+ kfree(osb->vol_label);
+out_recovery_map:
+ kfree(osb->recovery_map);
+out:
+ kfree(osb);
+ sb->s_fs_info = NULL;
return status;
}
@@ -2378,6 +2267,7 @@ static int ocfs2_verify_volume(struct ocfs2_dinode *di,
struct ocfs2_blockcheck_stats *stats)
{
int status = -EAGAIN;
+ u32 blksz_bits;
if (memcmp(di->i_signature, OCFS2_SUPER_BLOCK_SIGNATURE,
strlen(OCFS2_SUPER_BLOCK_SIGNATURE)) == 0) {
@@ -2392,11 +2282,15 @@ static int ocfs2_verify_volume(struct ocfs2_dinode *di,
goto out;
}
status = -EINVAL;
- if ((1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits)) != blksz) {
+ /* Acceptable block sizes are 512 bytes, 1K, 2K and 4K. */
+ blksz_bits = le32_to_cpu(di->id2.i_super.s_blocksize_bits);
+ if (blksz_bits < 9 || blksz_bits > 12) {
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);
+ "size bits: found %u, should be 9, 10, 11, or 12\n",
+ blksz_bits);
+ } else if ((1 << blksz_bits) != blksz) {
+ mlog(ML_ERROR, "found superblock with incorrect block "
+ "size: found %u, should be %u\n", 1 << blksz_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) !=
@@ -2414,8 +2308,8 @@ static int ocfs2_verify_volume(struct ocfs2_dinode *di,
(unsigned long long)bh->b_blocknr);
} 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));
+ mlog(ML_ERROR, "bad cluster size bit found: %u\n",
+ 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)) {
@@ -2448,7 +2342,7 @@ static int ocfs2_check_volume(struct ocfs2_super *osb)
* ourselves. */
/* Init our journal object. */
- status = ocfs2_journal_init(osb->journal, &dirty);
+ status = ocfs2_journal_init(osb, &dirty);
if (status < 0) {
mlog(ML_ERROR, "Could not initialize journal!\n");
goto finally;
@@ -2483,6 +2377,15 @@ static int ocfs2_check_volume(struct ocfs2_super *osb)
goto finally;
}
+ if (osb->s_mount_opt & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT)
+ jbd2_journal_set_features(osb->journal->j_journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+ JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+ else
+ jbd2_journal_clear_features(osb->journal->j_journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+ JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+
if (dirty) {
/* recover my local alloc if we didn't unmount cleanly. */
status = ocfs2_begin_local_alloc_recovery(osb,
@@ -2505,7 +2408,6 @@ static int ocfs2_check_volume(struct ocfs2_super *osb)
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;
}
@@ -2541,81 +2443,99 @@ static void ocfs2_delete_osb(struct ocfs2_super *osb)
{
/* This function assumes that the caller has the main osb resource */
+ /* ocfs2_initializer_super have already created this workqueue */
+ if (osb->ocfs2_wq)
+ destroy_workqueue(osb->ocfs2_wq);
+
ocfs2_free_slot_info(osb);
kfree(osb->osb_orphan_wipes);
kfree(osb->slot_recovery_generations);
/* FIXME
* This belongs in journal shutdown, but because we have to
- * allocate osb->journal at the start of ocfs2_initialize_osb(),
+ * allocate osb->journal at the middle of ocfs2_initialize_super(),
* we free it here.
*/
kfree(osb->journal);
kfree(osb->local_alloc_copy);
kfree(osb->uuid_str);
+ kfree(osb->vol_label);
ocfs2_put_dlm_debug(osb->osb_dlm_debug);
memset(osb, 0, sizeof(struct ocfs2_super));
}
-/* 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)
+/* Depending on the mount option passed, perform one of the following:
+ * Put OCFS2 into a readonly state (default)
+ * Return EIO so that only the process errs
+ * Fix the error as if fsck.ocfs2 -y
+ * panic
+ */
+static int ocfs2_handle_error(struct super_block *sb)
{
struct ocfs2_super *osb = OCFS2_SB(sb);
+ int rv = 0;
+
+ ocfs2_set_osb_flag(osb, OCFS2_OSB_ERROR_FS);
+ pr_crit("On-disk corruption discovered. "
+ "Please run fsck.ocfs2 once the filesystem is unmounted.\n");
- if (osb->s_mount_opt & OCFS2_MOUNT_ERRORS_PANIC)
+ if (osb->s_mount_opt & OCFS2_MOUNT_ERRORS_PANIC) {
panic("OCFS2: (device %s): panic forced after error\n",
sb->s_id);
+ } else if (osb->s_mount_opt & OCFS2_MOUNT_ERRORS_CONT) {
+ pr_crit("OCFS2: Returning error to the calling process.\n");
+ rv = -EIO;
+ } else { /* default option */
+ rv = -EROFS;
+ if (sb_rdonly(sb) && (ocfs2_is_soft_readonly(osb) || ocfs2_is_hard_readonly(osb)))
+ return rv;
- 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;
+ pr_crit("OCFS2: File system is now read-only.\n");
+ sb->s_flags |= SB_RDONLY;
+ ocfs2_set_ro_flag(osb, 0);
+ }
- 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);
+ return rv;
}
-static char error_buf[1024];
-
-void __ocfs2_error(struct super_block *sb,
- const char *function,
- const char *fmt, ...)
+int __ocfs2_error(struct super_block *sb, const char *function,
+ const char *fmt, ...)
{
+ struct va_format vaf;
va_list args;
va_start(args, fmt);
- vsnprintf(error_buf, sizeof(error_buf), fmt, args);
- va_end(args);
+ vaf.fmt = fmt;
+ vaf.va = &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);
+ printk(KERN_CRIT "OCFS2: ERROR (device %s): %s: %pV",
+ sb->s_id, function, &vaf);
- ocfs2_handle_error(sb);
+ va_end(args);
+
+ return 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,
+void __ocfs2_abort(struct super_block *sb, const char *function,
const char *fmt, ...)
{
+ struct va_format vaf;
va_list args;
va_start(args, fmt);
- vsnprintf(error_buf, sizeof(error_buf), fmt, args);
- va_end(args);
- printk(KERN_CRIT "OCFS2: abort (device %s): %s: %s\n",
- sb->s_id, function, error_buf);
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ printk(KERN_CRIT "OCFS2: abort (device %s): %s: %pV",
+ sb->s_id, function, &vaf);
+
+ va_end(args);
/* We don't have the cluster support yet to go straight to
* hard readonly in here. Until then, we want to keep
generated by cgit (git 2.34.1) at 2025-12-25 05:29:19 +0000