#include <linux/parser.h>
#include <linux/crc32.h>
#include <linux/debugfs.h>
-
-#include <cluster/nodemanager.h>
+#include <linux/mount.h>
+#include <linux/seq_file.h>
+#include <linux/quotaops.h>
#define MLOG_MASK_PREFIX ML_SUPER
#include <cluster/masklog.h>
#include "ocfs1_fs_compat.h"
#include "alloc.h"
+#include "blockcheck.h"
#include "dlmglue.h"
#include "export.h"
#include "extent_map.h"
#include "sysfile.h"
#include "uptodate.h"
#include "ver.h"
-#include "vote.h"
+#include "xattr.h"
+#include "quota.h"
#include "buffer_head_io.h"
-/*
- * Globals
- */
-static spinlock_t ocfs2_globals_lock = SPIN_LOCK_UNLOCKED;
-
-static u32 osb_id; /* Keeps track of next available OSB Id */
-
-static kmem_cache_t *ocfs2_inode_cachep = NULL;
-
-kmem_cache_t *ocfs2_lock_cache = NULL;
+static struct kmem_cache *ocfs2_inode_cachep = NULL;
+struct kmem_cache *ocfs2_dquot_cachep;
+struct kmem_cache *ocfs2_qf_chunk_cachep;
/* OCFS2 needs to schedule several differnt types of work which
* require cluster locking, disk I/O, recovery waits, etc. Since these
MODULE_AUTHOR("Oracle");
MODULE_LICENSE("GPL");
+struct mount_options
+{
+ unsigned long commit_interval;
+ unsigned long mount_opt;
+ unsigned int atime_quantum;
+ signed short slot;
+ unsigned int localalloc_opt;
+ char cluster_stack[OCFS2_STACK_LABEL_LEN + 1];
+};
+
static int ocfs2_parse_options(struct super_block *sb, char *options,
- unsigned long *mount_opt, int is_remount);
+ struct mount_options *mopt,
+ int is_remount);
+static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt);
static void ocfs2_put_super(struct super_block *sb);
static int ocfs2_mount_volume(struct super_block *sb);
static int ocfs2_remount(struct super_block *sb, int *flags, char *data);
static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb);
static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb);
-static int ocfs2_release_system_inodes(struct ocfs2_super *osb);
-static int ocfs2_fill_local_node_info(struct ocfs2_super *osb);
+static void ocfs2_release_system_inodes(struct ocfs2_super *osb);
static int ocfs2_check_volume(struct ocfs2_super *osb);
static int ocfs2_verify_volume(struct ocfs2_dinode *di,
struct buffer_head *bh,
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);
+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 unsigned long long ocfs2_max_file_offset(unsigned int blockshift);
-
-static struct super_operations ocfs2_sops = {
+static const struct super_operations ocfs2_sops = {
.statfs = ocfs2_statfs,
.alloc_inode = ocfs2_alloc_inode,
.destroy_inode = ocfs2_destroy_inode,
.write_super = ocfs2_write_super,
.put_super = ocfs2_put_super,
.remount_fs = ocfs2_remount,
+ .show_options = ocfs2_show_options,
+ .quota_read = ocfs2_quota_read,
+ .quota_write = ocfs2_quota_write,
};
enum {
Opt_hb_local,
Opt_data_ordered,
Opt_data_writeback,
+ Opt_atime_quantum,
+ Opt_slot,
+ Opt_commit,
+ Opt_localalloc,
+ Opt_localflocks,
+ Opt_stack,
+ Opt_user_xattr,
+ Opt_nouser_xattr,
+ Opt_inode64,
+ Opt_acl,
+ Opt_noacl,
+ Opt_usrquota,
+ Opt_grpquota,
Opt_err,
};
-static match_table_t tokens = {
+static const match_table_t tokens = {
{Opt_barrier, "barrier=%u"},
{Opt_err_panic, "errors=panic"},
{Opt_err_ro, "errors=remount-ro"},
{Opt_hb_local, OCFS2_HB_LOCAL},
{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_err, NULL}
};
+#ifdef CONFIG_DEBUG_FS
+static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
+{
+ int out = 0;
+ int i;
+ struct ocfs2_cluster_connection *cconn = osb->cconn;
+ struct ocfs2_recovery_map *rm = osb->recovery_map;
+
+ out += snprintf(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,
+ "%10s => State: %d Flags: 0x%lX\n", "Volume",
+ atomic_read(&osb->vol_state), osb->osb_flags);
+
+ out += snprintf(buf + out, len - out,
+ "%10s => Block: %lu Cluster: %d\n", "Sizes",
+ osb->sb->s_blocksize, osb->s_clustersize);
+
+ out += snprintf(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,
+ "%10s => Opts: 0x%lX AtimeQuanta: %u\n", "Mount",
+ osb->s_mount_opt, osb->s_atime_quantum);
+
+ out += snprintf(buf + out, len - out,
+ "%10s => Stack: %s Name: %*s Version: %d.%d\n",
+ "Cluster",
+ (*osb->osb_cluster_stack == '\0' ?
+ "o2cb" : osb->osb_cluster_stack),
+ cconn->cc_namelen, cconn->cc_name,
+ cconn->cc_version.pv_major, cconn->cc_version.pv_minor);
+
+ spin_lock(&osb->dc_task_lock);
+ out += snprintf(buf + out, len - out,
+ "%10s => Pid: %d Count: %lu WakeSeq: %lu "
+ "WorkSeq: %lu\n", "DownCnvt",
+ task_pid_nr(osb->dc_task), osb->blocked_lock_count,
+ osb->dc_wake_sequence, osb->dc_work_sequence);
+ spin_unlock(&osb->dc_task_lock);
+
+ spin_lock(&osb->osb_lock);
+ out += snprintf(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");
+ else {
+ for (i = 0; i < rm->rm_used; i++)
+ out += snprintf(buf + out, len - out, " %d",
+ rm->rm_entries[i]);
+ out += snprintf(buf + out, len - out, "\n");
+ }
+ spin_unlock(&osb->osb_lock);
+
+ out += snprintf(buf + out, len - out,
+ "%10s => Pid: %d Interval: %lu Needs: %d\n", "Commit",
+ task_pid_nr(osb->commit_task), osb->osb_commit_interval,
+ atomic_read(&osb->needs_checkpoint));
+
+ out += snprintf(buf + out, len - out,
+ "%10s => State: %d NumTxns: %d TxnId: %lu\n",
+ "Journal", osb->journal->j_state,
+ atomic_read(&osb->journal->j_num_trans),
+ osb->journal->j_trans_id);
+
+ out += snprintf(buf + out, len - out,
+ "%10s => GlobalAllocs: %d LocalAllocs: %d "
+ "SubAllocs: %d LAWinMoves: %d SAExtends: %d\n",
+ "Stats",
+ atomic_read(&osb->alloc_stats.bitmap_data),
+ atomic_read(&osb->alloc_stats.local_data),
+ atomic_read(&osb->alloc_stats.bg_allocs),
+ atomic_read(&osb->alloc_stats.moves),
+ atomic_read(&osb->alloc_stats.bg_extends));
+
+ out += snprintf(buf + out, len - out,
+ "%10s => State: %u Descriptor: %llu Size: %u bits "
+ "Default: %u bits\n",
+ "LocalAlloc", osb->local_alloc_state,
+ (unsigned long long)osb->la_last_gd,
+ osb->local_alloc_bits, osb->local_alloc_default_bits);
+
+ spin_lock(&osb->osb_lock);
+ out += snprintf(buf + out, len - out,
+ "%10s => Slot: %d NumStolen: %d\n", "Steal",
+ osb->s_inode_steal_slot,
+ atomic_read(&osb->s_num_inodes_stolen));
+ spin_unlock(&osb->osb_lock);
+
+ out += snprintf(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,
+ "%10s %c %3d %10d\n",
+ " ",
+ (i == osb->slot_num ? '*' : ' '),
+ i, osb->slot_recovery_generations[i]);
+ }
+
+ return out;
+}
+
+static int ocfs2_osb_debug_open(struct inode *inode, struct file *file)
+{
+ struct ocfs2_super *osb = inode->i_private;
+ char *buf = NULL;
+
+ buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!buf)
+ goto bail;
+
+ i_size_write(inode, ocfs2_osb_dump(osb, buf, PAGE_SIZE));
+
+ file->private_data = buf;
+
+ return 0;
+bail:
+ return -ENOMEM;
+}
+
+static int ocfs2_debug_release(struct inode *inode, struct file *file)
+{
+ kfree(file->private_data);
+ return 0;
+}
+
+static ssize_t ocfs2_debug_read(struct file *file, char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ return simple_read_from_buffer(buf, nbytes, ppos, file->private_data,
+ i_size_read(file->f_mapping->host));
+}
+#else
+static int ocfs2_osb_debug_open(struct inode *inode, struct file *file)
+{
+ return 0;
+}
+static int ocfs2_debug_release(struct inode *inode, struct file *file)
+{
+ return 0;
+}
+static ssize_t ocfs2_debug_read(struct file *file, char __user *buf,
+ size_t nbytes, loff_t *ppos)
+{
+ return 0;
+}
+#endif /* CONFIG_DEBUG_FS */
+
+static struct file_operations ocfs2_osb_debug_fops = {
+ .open = ocfs2_osb_debug_open,
+ .release = ocfs2_debug_release,
+ .read = ocfs2_debug_read,
+ .llseek = generic_file_llseek,
+};
+
/*
* write_super and sync_fs ripped right out of ext3.
*/
static int ocfs2_sync_fs(struct super_block *sb, int wait)
{
- int status = 0;
+ int status;
tid_t target;
struct ocfs2_super *osb = OCFS2_SB(sb);
ocfs2_schedule_truncate_log_flush(osb, 0);
}
- if (journal_start_commit(OCFS2_SB(sb)->journal->j_journal, &target)) {
+ if (jbd2_journal_start_commit(OCFS2_SB(sb)->journal->j_journal,
+ &target)) {
if (wait)
- log_wait_commit(OCFS2_SB(sb)->journal->j_journal,
- target);
+ jbd2_log_wait_commit(OCFS2_SB(sb)->journal->j_journal,
+ target);
}
return 0;
}
+static int ocfs2_need_system_inode(struct ocfs2_super *osb, int ino)
+{
+ if (!OCFS2_HAS_RO_COMPAT_FEATURE(osb->sb, OCFS2_FEATURE_RO_COMPAT_USRQUOTA)
+ && (ino == USER_QUOTA_SYSTEM_INODE
+ || ino == LOCAL_USER_QUOTA_SYSTEM_INODE))
+ return 0;
+ if (!OCFS2_HAS_RO_COMPAT_FEATURE(osb->sb, OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)
+ && (ino == GROUP_QUOTA_SYSTEM_INODE
+ || ino == LOCAL_GROUP_QUOTA_SYSTEM_INODE))
+ return 0;
+ return 1;
+}
+
static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb)
{
struct inode *new = NULL;
mlog_entry_void();
- new = ocfs2_iget(osb, osb->root_blkno);
+ new = ocfs2_iget(osb, osb->root_blkno, OCFS2_FI_FLAG_SYSFILE, 0);
if (IS_ERR(new)) {
status = PTR_ERR(new);
mlog_errno(status);
}
osb->root_inode = new;
- new = ocfs2_iget(osb, osb->system_dir_blkno);
+ new = ocfs2_iget(osb, osb->system_dir_blkno, OCFS2_FI_FLAG_SYSFILE, 0);
if (IS_ERR(new)) {
status = PTR_ERR(new);
mlog_errno(status);
for (i = OCFS2_FIRST_ONLINE_SYSTEM_INODE;
i <= OCFS2_LAST_GLOBAL_SYSTEM_INODE; i++) {
+ if (!ocfs2_need_system_inode(osb, i))
+ continue;
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
if (!new) {
ocfs2_release_system_inodes(osb);
for (i = OCFS2_LAST_GLOBAL_SYSTEM_INODE + 1;
i < NUM_SYSTEM_INODES;
i++) {
+ if (!ocfs2_need_system_inode(osb, i))
+ continue;
new = ocfs2_get_system_file_inode(osb, i, osb->slot_num);
if (!new) {
ocfs2_release_system_inodes(osb);
return status;
}
-static int ocfs2_release_system_inodes(struct ocfs2_super *osb)
+static void ocfs2_release_system_inodes(struct ocfs2_super *osb)
{
- int status = 0, i;
+ int i;
struct inode *inode;
mlog_entry_void();
osb->root_inode = NULL;
}
- mlog_exit(status);
- return status;
+ mlog_exit(0);
}
/* We're allocating fs objects, use GFP_NOFS */
{
struct ocfs2_inode_info *oi;
- oi = kmem_cache_alloc(ocfs2_inode_cachep, SLAB_NOFS);
+ oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS);
if (!oi)
return NULL;
+ jbd2_journal_init_jbd_inode(&oi->ip_jinode, &oi->vfs_inode);
return &oi->vfs_inode;
}
kmem_cache_free(ocfs2_inode_cachep, OCFS2_I(inode));
}
-/* From xfs_super.c:xfs_max_file_offset
- * Copyright (c) 2000-2004 Silicon Graphics, Inc.
- */
-static unsigned long long ocfs2_max_file_offset(unsigned int blockshift)
+static unsigned long long ocfs2_max_file_offset(unsigned int bbits,
+ unsigned int cbits)
{
- unsigned int pagefactor = 1;
- unsigned int bitshift = BITS_PER_LONG - 1;
-
- /* Figure out maximum filesize, on Linux this can depend on
- * the filesystem blocksize (on 32 bit platforms).
- * __block_prepare_write does this in an [unsigned] long...
- * page->index << (PAGE_CACHE_SHIFT - bbits)
- * So, for page sized blocks (4K on 32 bit platforms),
- * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is
- * (((u64)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1)
- * but for smaller blocksizes it is less (bbits = log2 bsize).
- * Note1: get_block_t takes a long (implicit cast from above)
- * Note2: The Large Block Device (LBD and HAVE_SECTOR_T) patch
- * can optionally convert the [unsigned] long from above into
- * an [unsigned] long long.
+ 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.
*/
#if BITS_PER_LONG == 32
# if defined(CONFIG_LBD)
- BUG_ON(sizeof(sector_t) != 8);
- pagefactor = PAGE_CACHE_SIZE;
- bitshift = BITS_PER_LONG;
+ BUILD_BUG_ON(sizeof(sector_t) != 8);
+ /*
+ * 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;
+ }
# else
- pagefactor = PAGE_CACHE_SIZE >> (PAGE_CACHE_SHIFT - blockshift);
+ /*
+ * 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
- return (((unsigned long long)pagefactor) << bitshift) - 1;
+ /*
+ * 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;
}
static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
{
int incompat_features;
int ret = 0;
- unsigned long parsed_options;
+ struct mount_options parsed_options;
struct ocfs2_super *osb = OCFS2_SB(sb);
if (!ocfs2_parse_options(sb, data, &parsed_options, 1)) {
}
if ((osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) !=
- (parsed_options & OCFS2_MOUNT_HB_LOCAL)) {
+ (parsed_options.mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
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 & 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;
}
+ /* 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)) {
+ 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)) {
+ /* Disable quota accounting before remounting RO */
+ if (*flags & MS_RDONLY) {
+ ret = ocfs2_susp_quotas(osb, 0);
+ if (ret < 0)
+ goto out;
+ }
/* Lock here so the check of HARD_RO and the potential
* setting of SOFT_RO is atomic. */
spin_lock(&osb->osb_lock);
}
unlock_osb:
spin_unlock(&osb->osb_lock);
+ /* Enable quota accounting after remounting RW */
+ if (!ret && !(*flags & MS_RDONLY)) {
+ if (sb_any_quota_suspended(sb))
+ ret = ocfs2_susp_quotas(osb, 1);
+ else
+ ret = ocfs2_enable_quotas(osb);
+ if (ret < 0) {
+ /* Return back changes... */
+ spin_lock(&osb->osb_lock);
+ sb->s_flags |= MS_RDONLY;
+ osb->osb_flags |= OCFS2_OSB_SOFT_RO;
+ spin_unlock(&osb->osb_lock);
+ goto out;
+ }
+ }
}
if (!ret) {
- if (!ocfs2_is_hard_readonly(osb))
- ocfs2_set_journal_params(osb);
-
/* Only save off the new mount options in case of a successful
* remount. */
- osb->s_mount_opt = parsed_options;
+ if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
+ parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
+ osb->s_mount_opt = parsed_options.mount_opt;
+ osb->s_atime_quantum = parsed_options.atime_quantum;
+ osb->preferred_slot = parsed_options.slot;
+ if (parsed_options.commit_interval)
+ osb->osb_commit_interval = parsed_options.commit_interval;
+
+ if (!ocfs2_is_hard_readonly(osb))
+ ocfs2_set_journal_params(osb);
}
out:
return ret;
struct buffer_head **bh,
int *sector_size)
{
- int status = 0, tmpstat;
+ int status, tmpstat;
struct ocfs1_vol_disk_hdr *hdr;
struct ocfs2_dinode *di;
int blksize;
return status;
}
+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 (ocfs2_userspace_stack(osb)) {
+ if (osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL) {
+ mlog(ML_ERROR, "Userspace stack expected, but "
+ "o2cb heartbeat arguments passed to mount\n");
+ return -EINVAL;
+ }
+ }
+
+ if (!(osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
+ if (!ocfs2_mount_local(osb) && !ocfs2_is_hard_readonly(osb) &&
+ !ocfs2_userspace_stack(osb)) {
+ mlog(ML_ERROR, "Heartbeat has to be started to mount "
+ "a read-write clustered device.\n");
+ return -EINVAL;
+ }
+ }
+
+ return 0;
+}
+
+/*
+ * If we're using a userspace stack, mount should have passed
+ * a name that matches the disk. If not, mount should not
+ * have passed a stack.
+ */
+static int ocfs2_verify_userspace_stack(struct ocfs2_super *osb,
+ struct mount_options *mopt)
+{
+ if (!ocfs2_userspace_stack(osb) && mopt->cluster_stack[0]) {
+ mlog(ML_ERROR,
+ "cluster stack passed to mount, but this filesystem "
+ "does not support it\n");
+ return -EINVAL;
+ }
+
+ if (ocfs2_userspace_stack(osb) &&
+ strncmp(osb->osb_cluster_stack, mopt->cluster_stack,
+ OCFS2_STACK_LABEL_LEN)) {
+ mlog(ML_ERROR,
+ "cluster stack passed to mount (\"%s\") does not "
+ "match the filesystem (\"%s\")\n",
+ mopt->cluster_stack,
+ osb->osb_cluster_stack);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+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};
+ int status = 0;
+
+ for (type = 0; type < MAXQUOTAS; type++) {
+ if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
+ continue;
+ if (unsuspend)
+ status = vfs_quota_enable(
+ sb_dqopt(sb)->files[type],
+ type, QFMT_OCFS2,
+ DQUOT_SUSPENDED);
+ else
+ status = vfs_quota_disable(sb, type,
+ DQUOT_SUSPENDED);
+ if (status < 0)
+ break;
+ }
+ if (status < 0)
+ mlog(ML_ERROR, "Failed to suspend/unsuspend quotas on "
+ "remount (error = %d).\n", status);
+ return status;
+}
+
+static int ocfs2_enable_quotas(struct ocfs2_super *osb)
+{
+ struct inode *inode[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,
+ 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++) {
+ if (!OCFS2_HAS_RO_COMPAT_FEATURE(sb, feature[type]))
+ continue;
+ inode[type] = ocfs2_get_system_file_inode(osb, ino[type],
+ osb->slot_num);
+ if (!inode[type]) {
+ status = -ENOENT;
+ goto out_quota_off;
+ }
+ status = vfs_quota_enable(inode[type], type, QFMT_OCFS2,
+ DQUOT_USAGE_ENABLED);
+ if (status < 0)
+ goto out_quota_off;
+ }
+
+ for (type = 0; type < MAXQUOTAS; type++)
+ iput(inode[type]);
+ return 0;
+out_quota_off:
+ ocfs2_disable_quotas(osb);
+ for (type = 0; type < MAXQUOTAS; type++)
+ iput(inode[type]);
+ mlog_errno(status);
+ return status;
+}
+
+static void ocfs2_disable_quotas(struct ocfs2_super *osb)
+{
+ int type;
+ struct inode *inode;
+ struct super_block *sb = osb->sb;
+
+ /* 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++) {
+ if (!sb_has_quota_loaded(sb, type))
+ continue;
+ 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 */
+ vfs_quota_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,
+ char *path, int remount)
+{
+ 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;
+
+ if (remount)
+ return 0; /* Just ignore it has been handled in
+ * ocfs2_remount() */
+ return vfs_quota_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, int remount)
+{
+ if (remount)
+ return 0; /* Ignore now and handle later in
+ * ocfs2_remount() */
+ return vfs_quota_disable(sb, type, DQUOT_LIMITS_ENABLED);
+}
+
+static struct quotactl_ops ocfs2_quotactl_ops = {
+ .quota_on = ocfs2_quota_on,
+ .quota_off = ocfs2_quota_off,
+ .quota_sync = vfs_quota_sync,
+ .get_info = vfs_get_dqinfo,
+ .set_info = vfs_set_dqinfo,
+ .get_dqblk = vfs_get_dqblk,
+ .set_dqblk = vfs_set_dqblk,
+};
+
static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
{
struct dentry *root;
int status, sector_size;
- unsigned long parsed_opt;
+ struct mount_options parsed_options;
struct inode *inode = NULL;
struct ocfs2_super *osb = NULL;
struct buffer_head *bh = NULL;
+ char nodestr[8];
mlog_entry("%p, %p, %i", sb, data, silent);
- /* for now we only have one cluster/node, make sure we see it
- * in the heartbeat universe */
- if (!o2hb_check_local_node_heartbeating()) {
+ if (!ocfs2_parse_options(sb, data, &parsed_options, 0)) {
status = -EINVAL;
goto read_super_error;
}
brelse(bh);
bh = NULL;
- if (!ocfs2_parse_options(sb, data, &parsed_opt, 0)) {
+ if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
+ parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
+
+ osb->s_mount_opt = parsed_options.mount_opt;
+ osb->s_atime_quantum = parsed_options.atime_quantum;
+ osb->preferred_slot = parsed_options.slot;
+ osb->osb_commit_interval = parsed_options.commit_interval;
+ osb->local_alloc_default_bits = ocfs2_megabytes_to_clusters(sb, parsed_options.localalloc_opt);
+ osb->local_alloc_bits = osb->local_alloc_default_bits;
+ if (osb->s_mount_opt & OCFS2_MOUNT_USRQUOTA &&
+ !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
status = -EINVAL;
+ mlog(ML_ERROR, "User quotas were requested, but this "
+ "filesystem does not have the feature enabled.\n");
+ goto read_super_error;
+ }
+ if (osb->s_mount_opt & OCFS2_MOUNT_GRPQUOTA &&
+ !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
+ status = -EINVAL;
+ mlog(ML_ERROR, "Group quotas were requested, but this "
+ "filesystem does not have the feature enabled.\n");
goto read_super_error;
}
- osb->s_mount_opt = parsed_opt;
+
+ status = ocfs2_verify_userspace_stack(osb, &parsed_options);
+ if (status)
+ goto read_super_error;
sb->s_magic = OCFS2_SUPER_MAGIC;
+ sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+ ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
+
/* Hard readonly mode only if: bdev_read_only, MS_RDONLY,
* heartbeat=none */
if (bdev_read_only(sb->s_bdev)) {
}
if (!ocfs2_is_hard_readonly(osb)) {
- /* If this isn't a hard readonly mount, then we need
- * to make sure that heartbeat is in a valid state,
- * and that we mark ourselves soft readonly is -oro
- * was specified. */
- if (!(osb->s_mount_opt & OCFS2_MOUNT_HB_LOCAL)) {
- mlog(ML_ERROR, "No heartbeat for device (%s)\n",
- sb->s_id);
- status = -EINVAL;
- goto read_super_error;
- }
-
if (sb->s_flags & MS_RDONLY)
ocfs2_set_ro_flag(osb, 0);
}
+ status = ocfs2_verify_heartbeat(osb);
+ if (status < 0) {
+ mlog_errno(status);
+ goto read_super_error;
+ }
+
osb->osb_debug_root = debugfs_create_dir(osb->uuid_str,
ocfs2_debugfs_root);
if (!osb->osb_debug_root) {
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;
+ }
+
status = ocfs2_mount_volume(sb);
if (osb->root_inode)
inode = igrab(osb->root_inode);
ocfs2_complete_mount_recovery(osb);
- printk("ocfs2: Mounting device (%u,%u) on (node %d, slot %d) with %s "
- "data mode.\n",
- MAJOR(sb->s_dev), MINOR(sb->s_dev), osb->node_num,
- osb->slot_num,
+ if (ocfs2_mount_local(osb))
+ snprintf(nodestr, sizeof(nodestr), "local");
+ else
+ snprintf(nodestr, sizeof(nodestr), "%u", osb->node_num);
+
+ printk(KERN_INFO "ocfs2: Mounting device (%s) on (node %s, slot %d) "
+ "with %s data mode.\n",
+ osb->dev_str, nodestr, osb->slot_num,
osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK ? "writeback" :
"ordered");
atomic_set(&osb->vol_state, VOLUME_MOUNTED);
wake_up(&osb->osb_mount_event);
+ /* 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)) {
+ status = ocfs2_enable_quotas(osb);
+ if (status < 0) {
+ /* We have to err-out specially here because
+ * s_root is already set */
+ mlog_errno(status);
+ atomic_set(&osb->vol_state, VOLUME_DISABLED);
+ wake_up(&osb->osb_mount_event);
+ mlog_exit(status);
+ return status;
+ }
+ }
+
+ ocfs2_complete_quota_recovery(osb);
+
+ /* Now we wake up again for processes waiting for quotas */
+ atomic_set(&osb->vol_state, VOLUME_MOUNTED_QUOTAS);
+ wake_up(&osb->osb_mount_event);
+
mlog_exit(status);
return status;
read_super_error:
- if (bh != NULL)
- brelse(bh);
+ brelse(bh);
if (inode)
iput(inode);
.kill_sb = kill_block_super, /* set to the generic one
* right now, but do we
* need to change that? */
- .fs_flags = FS_REQUIRES_DEV,
+ .fs_flags = FS_REQUIRES_DEV|FS_RENAME_DOES_D_MOVE,
.next = NULL
};
static int ocfs2_parse_options(struct super_block *sb,
char *options,
- unsigned long *mount_opt,
+ struct mount_options *mopt,
int is_remount)
{
int status;
mlog_entry("remount: %d, options: \"%s\"\n", is_remount,
options ? options : "(none)");
- *mount_opt = 0;
+ mopt->commit_interval = 0;
+ mopt->mount_opt = 0;
+ mopt->atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
+ mopt->slot = OCFS2_INVALID_SLOT;
+ mopt->localalloc_opt = OCFS2_DEFAULT_LOCAL_ALLOC_SIZE;
+ mopt->cluster_stack[0] = '\0';
if (!options) {
status = 1;
token = match_token(p, tokens, args);
switch (token) {
case Opt_hb_local:
- *mount_opt |= OCFS2_MOUNT_HB_LOCAL;
+ mopt->mount_opt |= OCFS2_MOUNT_HB_LOCAL;
break;
case Opt_hb_none:
- *mount_opt &= ~OCFS2_MOUNT_HB_LOCAL;
+ mopt->mount_opt &= ~OCFS2_MOUNT_HB_LOCAL;
break;
case Opt_barrier:
if (match_int(&args[0], &option)) {
goto bail;
}
if (option)
- *mount_opt |= OCFS2_MOUNT_BARRIER;
+ mopt->mount_opt |= OCFS2_MOUNT_BARRIER;
else
- *mount_opt &= ~OCFS2_MOUNT_BARRIER;
+ mopt->mount_opt &= ~OCFS2_MOUNT_BARRIER;
break;
case Opt_intr:
- *mount_opt &= ~OCFS2_MOUNT_NOINTR;
+ mopt->mount_opt &= ~OCFS2_MOUNT_NOINTR;
break;
case Opt_nointr:
- *mount_opt |= OCFS2_MOUNT_NOINTR;
+ mopt->mount_opt |= OCFS2_MOUNT_NOINTR;
break;
case Opt_err_panic:
- *mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
+ mopt->mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
break;
case Opt_err_ro:
- *mount_opt &= ~OCFS2_MOUNT_ERRORS_PANIC;
+ mopt->mount_opt &= ~OCFS2_MOUNT_ERRORS_PANIC;
break;
case Opt_data_ordered:
- *mount_opt &= ~OCFS2_MOUNT_DATA_WRITEBACK;
+ mopt->mount_opt &= ~OCFS2_MOUNT_DATA_WRITEBACK;
break;
case Opt_data_writeback:
- *mount_opt |= OCFS2_MOUNT_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:
+ 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 && (option <= ocfs2_local_alloc_size(sb) * 8))
+ 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';
+ break;
+ case Opt_inode64:
+ mopt->mount_opt |= OCFS2_MOUNT_INODE64;
+ break;
+ case Opt_usrquota:
+ /* We check only on remount, otherwise features
+ * aren't yet initialized. */
+ if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
+ mlog(ML_ERROR, "User quota requested but "
+ "filesystem feature is not set\n");
+ status = 0;
+ goto bail;
+ }
+ mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA;
+ break;
+ case Opt_grpquota:
+ if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+ OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
+ mlog(ML_ERROR, "Group quota requested but "
+ "filesystem feature is not set\n");
+ status = 0;
+ goto bail;
+ }
+ mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA;
+ break;
+#ifdef CONFIG_OCFS2_FS_POSIX_ACL
+ case Opt_acl:
+ mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+ break;
+ case Opt_noacl:
+ mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
break;
+#else
+ case Opt_acl:
+ case Opt_noacl:
+ printk(KERN_INFO "ocfs2 (no)acl options not supported\n");
+ break;
+#endif
default:
mlog(ML_ERROR,
"Unrecognized mount option \"%s\" "
return status;
}
+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;
+ unsigned int local_alloc_megs;
+
+ 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);
+
+ if (osb->osb_commit_interval)
+ seq_printf(s, ",commit=%u",
+ (unsigned) (osb->osb_commit_interval / HZ));
+
+ local_alloc_megs = osb->local_alloc_bits >> (20 - osb->s_clustersize_bits);
+ if (local_alloc_megs != OCFS2_DEFAULT_LOCAL_ALLOC_SIZE)
+ seq_printf(s, ",localalloc=%d", local_alloc_megs);
+
+ if (opts & OCFS2_MOUNT_LOCALFLOCKS)
+ seq_printf(s, ",localflocks,");
+
+ if (osb->osb_cluster_stack[0])
+ seq_printf(s, ",cluster_stack=%.*s", OCFS2_STACK_LABEL_LEN,
+ osb->osb_cluster_stack);
+ if (opts & OCFS2_MOUNT_USRQUOTA)
+ seq_printf(s, ",usrquota");
+ if (opts & OCFS2_MOUNT_GRPQUOTA)
+ seq_printf(s, ",grpquota");
+
+ if (opts & OCFS2_MOUNT_NOUSERXATTR)
+ seq_printf(s, ",nouser_xattr");
+ else
+ seq_printf(s, ",user_xattr");
+
+ if (opts & OCFS2_MOUNT_INODE64)
+ seq_printf(s, ",inode64");
+
+#ifdef CONFIG_OCFS2_FS_POSIX_ACL
+ if (opts & OCFS2_MOUNT_POSIX_ACL)
+ seq_printf(s, ",acl");
+ else
+ seq_printf(s, ",noacl");
+#endif
+
+ return 0;
+}
+
static int __init ocfs2_init(void)
{
int status;
ocfs2_print_version();
- if (init_ocfs2_extent_maps())
- return -ENOMEM;
-
status = init_ocfs2_uptodate_cache();
if (status < 0) {
mlog_errno(status);
goto leave;
}
- spin_lock(&ocfs2_globals_lock);
- osb_id = 0;
- spin_unlock(&ocfs2_globals_lock);
-
ocfs2_debugfs_root = debugfs_create_dir("ocfs2", NULL);
if (!ocfs2_debugfs_root) {
status = -EFAULT;
mlog(ML_ERROR, "Unable to create ocfs2 debugfs root.\n");
}
+ status = ocfs2_quota_setup();
+ if (status)
+ goto leave;
+
+ ocfs2_set_locking_protocol();
+
+ status = register_quota_format(&ocfs2_quota_format);
leave:
if (status < 0) {
+ ocfs2_quota_shutdown();
ocfs2_free_mem_caches();
exit_ocfs2_uptodate_cache();
- exit_ocfs2_extent_maps();
}
mlog_exit(status);
{
mlog_entry_void();
+ ocfs2_quota_shutdown();
+
if (ocfs2_wq) {
flush_workqueue(ocfs2_wq);
destroy_workqueue(ocfs2_wq);
}
+ unregister_quota_format(&ocfs2_quota_format);
+
debugfs_remove(ocfs2_debugfs_root);
ocfs2_free_mem_caches();
unregister_filesystem(&ocfs2_fs_type);
- exit_ocfs2_extent_maps();
-
exit_ocfs2_uptodate_cache();
mlog_exit_void();
goto bail;
}
- status = ocfs2_meta_lock(inode, NULL, &bh, 0);
+ status = ocfs2_inode_lock(inode, &bh, 0);
if (status < 0) {
mlog_errno(status);
goto bail;
brelse(bh);
- ocfs2_meta_unlock(inode, 0);
+ ocfs2_inode_unlock(inode, 0);
status = 0;
bail:
if (inode)
return status;
}
-static void ocfs2_inode_init_once(void *data,
- kmem_cache_t *cachep,
- unsigned long flags)
+static void ocfs2_inode_init_once(void *data)
{
struct ocfs2_inode_info *oi = data;
- if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
- SLAB_CTOR_CONSTRUCTOR) {
- 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_handle_list);
- INIT_LIST_HEAD(&oi->ip_io_markers);
- oi->ip_handle = NULL;
- oi->ip_created_trans = 0;
- oi->ip_last_trans = 0;
- oi->ip_dir_start_lookup = 0;
+ 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;
- init_rwsem(&oi->ip_alloc_sem);
- mutex_init(&oi->ip_io_mutex);
+ 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_blkno = 0ULL;
+ oi->ip_clusters = 0;
- ocfs2_lock_res_init_once(&oi->ip_rw_lockres);
- ocfs2_lock_res_init_once(&oi->ip_meta_lockres);
- ocfs2_lock_res_init_once(&oi->ip_data_lockres);
+ ocfs2_lock_res_init_once(&oi->ip_rw_lockres);
+ ocfs2_lock_res_init_once(&oi->ip_inode_lockres);
+ ocfs2_lock_res_init_once(&oi->ip_open_lockres);
- ocfs2_metadata_cache_init(&oi->vfs_inode);
+ ocfs2_metadata_cache_init(&oi->vfs_inode);
- inode_init_once(&oi->vfs_inode);
- }
+ inode_init_once(&oi->vfs_inode);
}
static int ocfs2_initialize_mem_caches(void)
0,
(SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
SLAB_MEM_SPREAD),
- ocfs2_inode_init_once, NULL);
- if (!ocfs2_inode_cachep)
- return -ENOMEM;
-
- ocfs2_lock_cache = kmem_cache_create("ocfs2_lock",
- sizeof(struct ocfs2_journal_lock),
- 0,
- SLAB_HWCACHE_ALIGN,
- NULL, NULL);
- if (!ocfs2_lock_cache)
+ 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),
+ NULL);
+ ocfs2_qf_chunk_cachep = kmem_cache_create("ocfs2_qf_chunk_cache",
+ sizeof(struct ocfs2_quota_chunk),
+ 0,
+ (SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD),
+ 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);
return -ENOMEM;
+ }
return 0;
}
{
if (ocfs2_inode_cachep)
kmem_cache_destroy(ocfs2_inode_cachep);
- if (ocfs2_lock_cache)
- kmem_cache_destroy(ocfs2_lock_cache);
-
ocfs2_inode_cachep = NULL;
- ocfs2_lock_cache = NULL;
+
+ if (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);
+ ocfs2_qf_chunk_cachep = NULL;
}
static int ocfs2_get_sector(struct super_block *sb,
unlock_buffer(*bh);
ll_rw_block(READ, 1, bh);
wait_on_buffer(*bh);
- return 0;
-}
-
-/* ocfs2 1.0 only allows one cluster and node identity per kernel image. */
-static int ocfs2_fill_local_node_info(struct ocfs2_super *osb)
-{
- int status;
-
- /* XXX hold a ref on the node while mounte? easy enough, if
- * desirable. */
- osb->node_num = o2nm_this_node();
- if (osb->node_num == O2NM_MAX_NODES) {
- mlog(ML_ERROR, "could not find this host's node number\n");
- status = -ENOENT;
- goto bail;
+ if (!buffer_uptodate(*bh)) {
+ mlog_errno(-EIO);
+ brelse(*bh);
+ *bh = NULL;
+ return -EIO;
}
- mlog(ML_NOTICE, "I am node %d\n", osb->node_num);
-
- status = 0;
-bail:
- return status;
+ return 0;
}
static int ocfs2_mount_volume(struct super_block *sb)
if (ocfs2_is_hard_readonly(osb))
goto leave;
- status = ocfs2_fill_local_node_info(osb);
- if (status < 0) {
- mlog_errno(status);
- goto leave;
- }
-
- status = ocfs2_register_hb_callbacks(osb);
- if (status < 0) {
- mlog_errno(status);
- goto leave;
- }
-
status = ocfs2_dlm_init(osb);
if (status < 0) {
mlog_errno(status);
goto leave;
}
- /* requires vote_thread to be running. */
- status = ocfs2_register_net_handlers(osb);
- if (status < 0) {
- mlog_errno(status);
- goto leave;
- }
-
status = ocfs2_super_lock(osb, 1);
if (status < 0) {
mlog_errno(status);
goto leave;
}
- ocfs2_populate_mounted_map(osb);
-
/* load all node-local system inodes */
status = ocfs2_init_local_system_inodes(osb);
if (status < 0) {
goto leave;
}
- /* This should be sent *after* we recovered our journal as it
- * will cause other nodes to unmark us as needing
- * recovery. However, we need to send it *before* dropping the
- * super block lock as otherwise their recovery threads might
- * try to clean us up while we're live! */
- status = ocfs2_request_mount_vote(osb);
- if (status < 0)
- mlog_errno(status);
+ if (ocfs2_mount_local(osb))
+ goto leave;
leave:
if (unlock_super)
return status;
}
-/* we can't grab the goofy sem lock from inside wait_event, so we use
- * memory barriers to make sure that we'll see the null task before
- * being woken up */
-static int ocfs2_recovery_thread_running(struct ocfs2_super *osb)
-{
- mb();
- return osb->recovery_thread_task != NULL;
-}
-
static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
{
- int tmp;
+ int tmp, hangup_needed = 0;
struct ocfs2_super *osb = NULL;
+ char nodestr[8];
mlog_entry("(0x%p)\n", sb);
osb = OCFS2_SB(sb);
BUG_ON(!osb);
+ debugfs_remove(osb->osb_ctxt);
+
+ ocfs2_disable_quotas(osb);
+
ocfs2_shutdown_local_alloc(osb);
ocfs2_truncate_log_shutdown(osb);
- /* disable any new recovery threads and wait for any currently
- * running ones to exit. Do this before setting the vol_state. */
- mutex_lock(&osb->recovery_lock);
- osb->disable_recovery = 1;
- mutex_unlock(&osb->recovery_lock);
- wait_event(osb->recovery_event, !ocfs2_recovery_thread_running(osb));
-
- /* At this point, we know that no more recovery threads can be
- * launched, so wait for any recovery completion work to
- * complete. */
- flush_workqueue(ocfs2_wq);
+ /* This will disable recovery and flush any recovery work. */
+ ocfs2_recovery_exit(osb);
ocfs2_journal_shutdown(osb);
ocfs2_sync_blockdev(sb);
- /* No dlm means we've failed during mount, so skip all the
- * steps which depended on that to complete. */
- if (osb->dlm) {
+ /* No cluster connection means we've failed during mount, so skip
+ * all the steps which depended on that to complete. */
+ if (osb->cconn) {
tmp = ocfs2_super_lock(osb, 1);
if (tmp < 0) {
mlog_errno(tmp);
return;
}
+ }
- tmp = ocfs2_request_umount_vote(osb);
- if (tmp < 0)
- mlog_errno(tmp);
-
- if (osb->slot_num != OCFS2_INVALID_SLOT)
- ocfs2_put_slot(osb);
+ if (osb->slot_num != OCFS2_INVALID_SLOT)
+ ocfs2_put_slot(osb);
+ if (osb->cconn)
ocfs2_super_unlock(osb, 1);
- }
ocfs2_release_system_inodes(osb);
- if (osb->dlm) {
- ocfs2_unregister_net_handlers(osb);
-
- ocfs2_dlm_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.
+ * 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)
+ hangup_needed = 1;
- ocfs2_clear_hb_callbacks(osb);
+ if (osb->cconn)
+ ocfs2_dlm_shutdown(osb, hangup_needed);
debugfs_remove(osb->osb_debug_root);
- if (!mnt_err)
- ocfs2_stop_heartbeat(osb);
+ if (hangup_needed)
+ ocfs2_cluster_hangup(osb->uuid_str, strlen(osb->uuid_str));
atomic_set(&osb->vol_state, VOLUME_DISMOUNTED);
- printk("ocfs2: Unmounting device (%u,%u) on (node %d)\n",
- MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev), osb->node_num);
+ if (ocfs2_mount_local(osb))
+ snprintf(nodestr, sizeof(nodestr), "local");
+ else
+ snprintf(nodestr, sizeof(nodestr), "%u", osb->node_num);
+
+ printk(KERN_INFO "ocfs2: Unmounting device (%s) on (node %s)\n",
+ osb->dev_str, nodestr);
ocfs2_delete_osb(osb);
kfree(osb);
BUG_ON(uuid_bytes != OCFS2_VOL_UUID_LEN);
- osb->uuid_str = kcalloc(1, OCFS2_VOL_UUID_LEN * 2 + 1, GFP_KERNEL);
+ osb->uuid_str = kzalloc(OCFS2_VOL_UUID_LEN * 2 + 1, GFP_KERNEL);
if (osb->uuid_str == NULL)
return -ENOMEM;
- memcpy(osb->uuid, uuid, OCFS2_VOL_UUID_LEN);
-
for (i = 0, ptr = osb->uuid_str; i < OCFS2_VOL_UUID_LEN; i++) {
/* print with null */
ret = snprintf(ptr, 3, "%02X", uuid[i]);
struct buffer_head *bh,
int sector_size)
{
- int status = 0;
- int i;
- struct ocfs2_dinode *di = NULL;
+ int status;
+ int i, cbits, bbits;
+ struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
struct inode *inode = NULL;
- struct buffer_head *bitmap_bh = NULL;
struct ocfs2_journal *journal;
__le32 uuid_net_key;
struct ocfs2_super *osb;
mlog_entry_void();
- osb = kcalloc(1, sizeof(struct ocfs2_super), GFP_KERNEL);
+ osb = kzalloc(sizeof(struct ocfs2_super), GFP_KERNEL);
if (!osb) {
status = -ENOMEM;
mlog_errno(status);
sb->s_fs_info = osb;
sb->s_op = &ocfs2_sops;
sb->s_export_op = &ocfs2_export_ops;
+ sb->s_qcop = &ocfs2_quotactl_ops;
+ sb->dq_op = &ocfs2_quota_operations;
+ sb->s_xattr = ocfs2_xattr_handlers;
+ sb->s_time_gran = 1;
sb->s_flags |= MS_NOATIME;
/* this is needed to support O_LARGEFILE */
- sb->s_maxbytes = ocfs2_max_file_offset(sb->s_blocksize_bits);
+ 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);
+
+ osb->osb_dx_mask = (1 << (cbits - bbits)) - 1;
+
+ for (i = 0; i < 3; i++)
+ osb->osb_dx_seed[i] = le32_to_cpu(di->id2.i_super.s_dx_seed[i]);
+ 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);
- osb->net_response_ids = 0;
- spin_lock_init(&osb->net_response_lock);
- INIT_LIST_HEAD(&osb->net_response_list);
-
- INIT_LIST_HEAD(&osb->osb_net_handlers);
- init_waitqueue_head(&osb->recovery_event);
- spin_lock_init(&osb->vote_task_lock);
- init_waitqueue_head(&osb->vote_event);
- osb->vote_work_sequence = 0;
- osb->vote_wake_sequence = 0;
+ spin_lock_init(&osb->dc_task_lock);
+ init_waitqueue_head(&osb->dc_event);
+ osb->dc_work_sequence = 0;
+ osb->dc_wake_sequence = 0;
INIT_LIST_HEAD(&osb->blocked_lock_list);
osb->blocked_lock_count = 0;
- INIT_LIST_HEAD(&osb->vote_list);
spin_lock_init(&osb->osb_lock);
+ spin_lock_init(&osb->osb_xattr_lock);
+ ocfs2_init_inode_steal_slot(osb);
atomic_set(&osb->alloc_stats.moves, 0);
atomic_set(&osb->alloc_stats.local_data, 0);
snprintf(osb->dev_str, sizeof(osb->dev_str), "%u,%u",
MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev));
- mutex_init(&osb->recovery_lock);
-
- osb->disable_recovery = 0;
- osb->recovery_thread_task = NULL;
+ status = ocfs2_recovery_init(osb);
+ if (status) {
+ mlog(ML_ERROR, "Unable to initialize recovery state\n");
+ mlog_errno(status);
+ goto bail;
+ }
init_waitqueue_head(&osb->checkpoint_event);
atomic_set(&osb->needs_checkpoint, 0);
- osb->node_num = O2NM_INVALID_NODE_NUM;
+ osb->s_atime_quantum = OCFS2_DEFAULT_ATIME_QUANTUM;
+
osb->slot_num = OCFS2_INVALID_SLOT;
+ osb->s_xattr_inline_size = le16_to_cpu(
+ di->id2.i_super.s_xattr_inline_size);
+
osb->local_alloc_state = OCFS2_LA_UNUSED;
osb->local_alloc_bh = NULL;
-
- ocfs2_setup_hb_callbacks(osb);
+ INIT_DELAYED_WORK(&osb->la_enable_wq, ocfs2_la_enable_worker);
init_waitqueue_head(&osb->osb_mount_event);
goto bail;
}
- osb->uuid = kmalloc(OCFS2_VOL_UUID_LEN, GFP_KERNEL);
- if (!osb->uuid) {
- mlog(ML_ERROR, "unable to alloc uuid\n");
- status = -ENOMEM;
- goto bail;
- }
-
- di = (struct ocfs2_dinode *)bh->b_data;
-
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",
status = -EINVAL;
goto bail;
}
- mlog(ML_NOTICE, "max_slots for this device: %u\n", osb->max_slots);
+ mlog(0, "max_slots for this device: %u\n", osb->max_slots);
+
+ osb->slot_recovery_generations =
+ kcalloc(osb->max_slots, sizeof(*osb->slot_recovery_generations),
+ GFP_KERNEL);
+ if (!osb->slot_recovery_generations) {
+ status = -ENOMEM;
+ mlog_errno(status);
+ goto bail;
+ }
init_waitqueue_head(&osb->osb_wipe_event);
osb->osb_orphan_wipes = kcalloc(osb->max_slots,
goto bail;
}
+ if (ocfs2_userspace_stack(osb)) {
+ 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;
+ }
+ } else {
+ /* The empty string is identical with classic tools that
+ * don't know about s_cluster_info. */
+ osb->osb_cluster_stack[0] = '\0';
+ }
+
get_random_bytes(&osb->s_next_generation, sizeof(u32));
/* FIXME
*/
/* initialize our journal structure */
- journal = kcalloc(1, sizeof(struct ocfs2_journal), GFP_KERNEL);
+ journal = kzalloc(sizeof(struct ocfs2_journal), GFP_KERNEL);
if (!journal) {
mlog(ML_ERROR, "unable to alloc journal\n");
status = -ENOMEM;
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, osb);
+ INIT_WORK(&journal->j_recovery_work, ocfs2_complete_recovery);
journal->j_state = OCFS2_JOURNAL_FREE;
+ INIT_WORK(&osb->dentry_lock_work, ocfs2_drop_dl_inodes);
+ osb->dentry_lock_list = NULL;
+
/* get some pseudo constants for clustersize bits */
osb->s_clustersize_bits =
le32_to_cpu(di->id2.i_super.s_clustersize_bits);
goto bail;
}
- memcpy(&uuid_net_key, &osb->uuid[i], sizeof(osb->net_key));
- osb->net_key = le32_to_cpu(uuid_net_key);
+ 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';
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);
+ osb->uuid_hash = le32_to_cpu(di->id2.i_super.s_uuid_hash);
mlog(0, "vol_label: %s\n", osb->vol_label);
mlog(0, "uuid: %s\n", osb->uuid_str);
mlog(0, "root_blkno=%llu, system_dir_blkno=%llu\n",
}
osb->bitmap_blkno = OCFS2_I(inode)->ip_blkno;
-
- status = ocfs2_read_block(osb, osb->bitmap_blkno, &bitmap_bh, 0,
- inode);
iput(inode);
- if (status < 0) {
- mlog_errno(status);
- goto bail;
- }
- di = (struct ocfs2_dinode *) bitmap_bh->b_data;
- osb->bitmap_cpg = le16_to_cpu(di->id2.i_chain.cl_cpg);
- osb->num_clusters = le32_to_cpu(di->id1.bitmap1.i_total);
- brelse(bitmap_bh);
- mlog(0, "cluster bitmap inode: %llu, clusters per group: %u\n",
- (unsigned long long)osb->bitmap_blkno, osb->bitmap_cpg);
+ osb->bitmap_cpg = ocfs2_group_bitmap_size(sb) * 8;
status = ocfs2_init_slot_info(osb);
if (status < 0) {
goto bail;
}
- /* Link this osb onto the global linked list of all osb structures. */
- /* The Global Link List is mainted for the whole driver . */
- spin_lock(&ocfs2_globals_lock);
- osb->osb_id = osb_id;
- if (osb_id < OCFS2_MAX_OSB_ID)
- osb_id++;
- else {
- mlog(ML_ERROR, "Too many volumes mounted\n");
- status = -ENOMEM;
- }
- spin_unlock(&ocfs2_globals_lock);
-
bail:
mlog_exit(status);
return status;
if (memcmp(di->i_signature, OCFS2_SUPER_BLOCK_SIGNATURE,
strlen(OCFS2_SUPER_BLOCK_SIGNATURE)) == 0) {
+ /* We have to do a raw check of the feature here */
+ if (le32_to_cpu(di->id2.i_super.s_feature_incompat) &
+ OCFS2_FEATURE_INCOMPAT_META_ECC) {
+ status = ocfs2_block_check_validate(bh->b_data,
+ bh->b_size,
+ &di->i_check);
+ if (status)
+ goto out;
+ }
status = -EINVAL;
if ((1 << le32_to_cpu(di->id2.i_super.s_blocksize_bits)) != blksz) {
mlog(ML_ERROR, "found superblock with incorrect block "
} else if (bh->b_blocknr != le64_to_cpu(di->i_blkno)) {
mlog(ML_ERROR, "bad block number on superblock: "
"found %llu, should be %llu\n",
- (unsigned long long)di->i_blkno,
+ (unsigned long long)le64_to_cpu(di->i_blkno),
(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) {
}
}
+out:
mlog_exit(status);
return status;
}
static int ocfs2_check_volume(struct ocfs2_super *osb)
{
- int status = 0;
+ int status;
int dirty;
+ int local;
struct ocfs2_dinode *local_alloc = NULL; /* only used if we
* recover
* ourselves. */
"recovering volume.\n");
}
+ local = ocfs2_mount_local(osb);
+
/* will play back anything left in the journal. */
- ocfs2_journal_load(osb->journal);
+ status = ocfs2_journal_load(osb->journal, local, dirty);
+ if (status < 0) {
+ mlog(ML_ERROR, "ocfs2 journal load failed! %d\n", status);
+ goto finally;
+ }
if (dirty) {
/* recover my local alloc if we didn't unmount cleanly. */
* 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);
+ goto finally;
+ }
+
+ status = ocfs2_compute_replay_slots(osb);
if (status < 0)
mlog_errno(status);
/* This function assumes that the caller has the main osb resource */
- if (osb->slot_info)
- ocfs2_free_slot_info(osb->slot_info);
+ 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_initalize_osb(),
va_list args;
va_start(args, fmt);
- vsprintf(error_buf, fmt, args);
+ vsnprintf(error_buf, sizeof(error_buf), fmt, args);
va_end(args);
/* Not using mlog here because we want to show the actual
va_list args;
va_start(args, fmt);
- vsprintf(error_buf, fmt, args);
+ vsnprintf(error_buf, sizeof(error_buf), fmt, args);
va_end(args);
printk(KERN_CRIT "OCFS2: abort (device %s): %s: %s\n",