/*
* Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved.
- * Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved.
+ * Copyright (C) 2004-2007 Red Hat, Inc. All rights reserved.
*
* This copyrighted material is made available to anyone wishing to use,
* modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License v.2.
+ * of the GNU General Public License version 2.
*/
#include <linux/sched.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/buffer_head.h>
-#include <linux/vmalloc.h>
#include <linux/blkdev.h>
#include <linux/kthread.h>
-#include <asm/semaphore.h>
+#include <linux/namei.h>
+#include <linux/mount.h>
+#include <linux/gfs2_ondisk.h>
+#include <linux/lm_interface.h>
#include "gfs2.h"
+#include "incore.h"
+#include "bmap.h"
#include "daemon.h"
#include "glock.h"
#include "glops.h"
#include "inode.h"
#include "lm.h"
#include "mount.h"
-#include "ops_export.h"
#include "ops_fstype.h"
+#include "ops_dentry.h"
#include "ops_super.h"
#include "recovery.h"
#include "rgrp.h"
#include "super.h"
-#include "unlinked.h"
#include "sys.h"
+#include "util.h"
+#include "log.h"
#define DO 0
#define UNDO 1
static struct gfs2_sbd *init_sbd(struct super_block *sb)
{
struct gfs2_sbd *sdp;
- unsigned int x;
- sdp = vmalloc(sizeof(struct gfs2_sbd));
+ sdp = kzalloc(sizeof(struct gfs2_sbd), GFP_KERNEL);
if (!sdp)
return NULL;
- memset(sdp, 0, sizeof(struct gfs2_sbd));
-
- set_v2sdp(sb, sdp);
+ sb->s_fs_info = sdp;
sdp->sd_vfs = sb;
gfs2_tune_init(&sdp->sd_tune);
- for (x = 0; x < GFS2_GL_HASH_SIZE; x++) {
- sdp->sd_gl_hash[x].hb_lock = RW_LOCK_UNLOCKED;
- INIT_LIST_HEAD(&sdp->sd_gl_hash[x].hb_list);
- }
INIT_LIST_HEAD(&sdp->sd_reclaim_list);
spin_lock_init(&sdp->sd_reclaim_lock);
init_waitqueue_head(&sdp->sd_reclaim_wq);
- mutex_init(&sdp->sd_invalidate_inodes_mutex);
mutex_init(&sdp->sd_inum_mutex);
spin_lock_init(&sdp->sd_statfs_spin);
- mutex_init(&sdp->sd_statfs_mutex);
spin_lock_init(&sdp->sd_rindex_spin);
mutex_init(&sdp->sd_rindex_mutex);
spin_lock_init(&sdp->sd_jindex_spin);
mutex_init(&sdp->sd_jindex_mutex);
- INIT_LIST_HEAD(&sdp->sd_unlinked_list);
- spin_lock_init(&sdp->sd_unlinked_spin);
- mutex_init(&sdp->sd_unlinked_mutex);
-
INIT_LIST_HEAD(&sdp->sd_quota_list);
spin_lock_init(&sdp->sd_quota_spin);
mutex_init(&sdp->sd_quota_mutex);
spin_lock_init(&sdp->sd_log_lock);
- init_waitqueue_head(&sdp->sd_log_trans_wq);
- init_waitqueue_head(&sdp->sd_log_flush_wq);
- INIT_LIST_HEAD(&sdp->sd_log_le_gl);
INIT_LIST_HEAD(&sdp->sd_log_le_buf);
INIT_LIST_HEAD(&sdp->sd_log_le_revoke);
INIT_LIST_HEAD(&sdp->sd_log_le_rg);
INIT_LIST_HEAD(&sdp->sd_log_le_databuf);
+ INIT_LIST_HEAD(&sdp->sd_log_le_ordered);
- INIT_LIST_HEAD(&sdp->sd_log_blks_list);
- init_waitqueue_head(&sdp->sd_log_blks_wait);
-
+ mutex_init(&sdp->sd_log_reserve_mutex);
INIT_LIST_HEAD(&sdp->sd_ail1_list);
INIT_LIST_HEAD(&sdp->sd_ail2_list);
- mutex_init(&sdp->sd_log_flush_lock);
- INIT_LIST_HEAD(&sdp->sd_log_flush_list);
+ init_rwsem(&sdp->sd_log_flush_lock);
+ atomic_set(&sdp->sd_log_in_flight, 0);
+ init_waitqueue_head(&sdp->sd_log_flush_wait);
INIT_LIST_HEAD(&sdp->sd_revoke_list);
return sdp;
}
-static void init_vfs(struct gfs2_sbd *sdp)
+static void init_vfs(struct super_block *sb, unsigned noatime)
{
- struct super_block *sb = sdp->sd_vfs;
+ struct gfs2_sbd *sdp = sb->s_fs_info;
sb->s_magic = GFS2_MAGIC;
sb->s_op = &gfs2_super_ops;
sb->s_export_op = &gfs2_export_ops;
+ sb->s_time_gran = 1;
sb->s_maxbytes = MAX_LFS_FILESIZE;
if (sb->s_flags & (MS_NOATIME | MS_NODIRATIME))
- set_bit(SDF_NOATIME, &sdp->sd_flags);
+ set_bit(noatime, &sdp->sd_flags);
/* Don't let the VFS update atimes. GFS2 handles this itself. */
sb->s_flags |= MS_NOATIME | MS_NODIRATIME;
-
- /* Set up the buffer cache and fill in some fake block size values
- to allow us to read-in the on-disk superblock. */
- sdp->sd_sb.sb_bsize = sb_min_blocksize(sb, GFS2_BASIC_BLOCK);
- sdp->sd_sb.sb_bsize_shift = sb->s_blocksize_bits;
- sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT;
- sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift;
}
static int init_names(struct gfs2_sbd *sdp, int silent)
{
- struct gfs2_sb *sb = NULL;
char *proto, *table;
int error = 0;
/* Try to autodetect */
if (!proto[0] || !table[0]) {
- struct buffer_head *bh;
- bh = sb_getblk(sdp->sd_vfs,
- GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift);
- lock_buffer(bh);
- clear_buffer_uptodate(bh);
- clear_buffer_dirty(bh);
- unlock_buffer(bh);
- ll_rw_block(READ, 1, &bh);
- wait_on_buffer(bh);
-
- if (!buffer_uptodate(bh)) {
- brelse(bh);
- return -EIO;
- }
-
- sb = kmalloc(sizeof(struct gfs2_sb), GFP_KERNEL);
- if (!sb) {
- brelse(bh);
- return -ENOMEM;
- }
- gfs2_sb_in(sb, bh->b_data);
- brelse(bh);
+ error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift);
+ if (error)
+ return error;
- error = gfs2_check_sb(sdp, sb, silent);
+ error = gfs2_check_sb(sdp, &sdp->sd_sb, silent);
if (error)
goto out;
if (!proto[0])
- proto = sb->sb_lockproto;
+ proto = sdp->sd_sb.sb_lockproto;
if (!table[0])
- table = sb->sb_locktable;
+ table = sdp->sd_sb.sb_locktable;
}
if (!table[0])
snprintf(sdp->sd_proto_name, GFS2_FSNAME_LEN, "%s", proto);
snprintf(sdp->sd_table_name, GFS2_FSNAME_LEN, "%s", table);
- out:
- kfree(sb);
+ table = sdp->sd_table_name;
+ while ((table = strchr(table, '/')))
+ *table = '_';
+out:
return error;
}
if (undo)
goto fail_trans;
- p = kthread_run(gfs2_scand, sdp, "gfs2_scand");
- error = IS_ERR(p);
- if (error) {
- fs_err(sdp, "can't start scand thread: %d\n", error);
- return error;
- }
- sdp->sd_scand_process = p;
-
for (sdp->sd_glockd_num = 0;
sdp->sd_glockd_num < sdp->sd_args.ar_num_glockd;
sdp->sd_glockd_num++) {
error = gfs2_glock_nq_num(sdp,
GFS2_LIVE_LOCK, &gfs2_nondisk_glops,
LM_ST_SHARED,
- LM_FLAG_NOEXP | GL_EXACT | GL_NEVER_RECURSE,
+ LM_FLAG_NOEXP | GL_EXACT,
&sdp->sd_live_gh);
if (error) {
fs_err(sdp, "can't acquire live glock: %d\n", error);
return 0;
- fail_trans:
+fail_trans:
gfs2_glock_put(sdp->sd_trans_gl);
-
- fail_rename:
+fail_rename:
gfs2_glock_put(sdp->sd_rename_gl);
-
- fail_live:
+fail_live:
gfs2_glock_dq_uninit(&sdp->sd_live_gh);
-
- fail_mount:
+fail_mount:
gfs2_glock_dq_uninit(mount_gh);
-
- fail:
+fail:
while (sdp->sd_glockd_num--)
kthread_stop(sdp->sd_glockd_process[sdp->sd_glockd_num]);
- kthread_stop(sdp->sd_scand_process);
-
return error;
}
-int gfs2_lookup_root(struct gfs2_sbd *sdp)
+static inline struct inode *gfs2_lookup_root(struct super_block *sb,
+ u64 no_addr)
{
- int error;
- struct gfs2_glock *gl;
- struct gfs2_inode *ip;
-
- error = gfs2_glock_get(sdp, sdp->sd_sb.sb_root_dir.no_addr,
- &gfs2_inode_glops, CREATE, &gl);
- if (!error) {
- error = gfs2_inode_get(gl, &sdp->sd_sb.sb_root_dir,
- CREATE, &ip);
- if (!error) {
- if (!error)
- gfs2_inode_min_init(ip, DT_DIR);
- sdp->sd_root_dir = gfs2_ip2v(ip);
- gfs2_inode_put(ip);
- }
- gfs2_glock_put(gl);
- }
-
- return error;
+ return gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0);
}
static int init_sb(struct gfs2_sbd *sdp, int silent, int undo)
{
struct super_block *sb = sdp->sd_vfs;
struct gfs2_holder sb_gh;
+ u64 no_addr;
struct inode *inode;
int error = 0;
if (undo) {
- iput(sdp->sd_master_dir);
+ if (sb->s_root) {
+ dput(sb->s_root);
+ sb->s_root = NULL;
+ }
return 0;
}
-
- error = gfs2_glock_nq_num(sdp,
- GFS2_SB_LOCK, &gfs2_meta_glops,
+
+ error = gfs2_glock_nq_num(sdp, GFS2_SB_LOCK, &gfs2_meta_glops,
LM_ST_SHARED, 0, &sb_gh);
if (error) {
fs_err(sdp, "can't acquire superblock glock: %d\n", error);
}
/* Set up the buffer cache and SB for real */
- error = -EINVAL;
if (sdp->sd_sb.sb_bsize < bdev_hardsect_size(sb->s_bdev)) {
+ error = -EINVAL;
fs_err(sdp, "FS block size (%u) is too small for device "
"block size (%u)\n",
sdp->sd_sb.sb_bsize, bdev_hardsect_size(sb->s_bdev));
goto out;
}
if (sdp->sd_sb.sb_bsize > PAGE_SIZE) {
+ error = -EINVAL;
fs_err(sdp, "FS block size (%u) is too big for machine "
"page size (%u)\n",
sdp->sd_sb.sb_bsize, (unsigned int)PAGE_SIZE);
goto out;
}
-
- /* Get rid of buffers from the original block size */
- sb_gh.gh_gl->gl_ops->go_inval(sb_gh.gh_gl, DIO_METADATA | DIO_DATA);
- sb_gh.gh_gl->gl_aspace->i_blkbits = sdp->sd_sb.sb_bsize_shift;
-
sb_set_blocksize(sb, sdp->sd_sb.sb_bsize);
/* Get the root inode */
- error = gfs2_lookup_root(sdp);
- if (error) {
+ no_addr = sdp->sd_sb.sb_root_dir.no_addr;
+ if (sb->s_type == &gfs2meta_fs_type)
+ no_addr = sdp->sd_sb.sb_master_dir.no_addr;
+ inode = gfs2_lookup_root(sb, no_addr);
+ if (IS_ERR(inode)) {
+ error = PTR_ERR(inode);
fs_err(sdp, "can't read in root inode: %d\n", error);
goto out;
}
- /* Get the root inode/dentry */
- inode = sdp->sd_root_dir;
- if (!inode) {
- fs_err(sdp, "can't get root inode\n");
- error = -ENOMEM;
- goto out_rooti;
- }
-
- igrab(inode);
sb->s_root = d_alloc_root(inode);
if (!sb->s_root) {
fs_err(sdp, "can't get root dentry\n");
error = -ENOMEM;
- goto out_rooti;
- }
-
+ iput(inode);
+ } else
+ sb->s_root->d_op = &gfs2_dops;
+
out:
gfs2_glock_dq_uninit(&sb_gh);
-
return error;
-out_rooti:
- iput(sdp->sd_root_dir);
- goto out;
+}
+
+/**
+ * map_journal_extents - create a reusable "extent" mapping from all logical
+ * blocks to all physical blocks for the given journal. This will save
+ * us time when writing journal blocks. Most journals will have only one
+ * extent that maps all their logical blocks. That's because gfs2.mkfs
+ * arranges the journal blocks sequentially to maximize performance.
+ * So the extent would map the first block for the entire file length.
+ * However, gfs2_jadd can happen while file activity is happening, so
+ * those journals may not be sequential. Less likely is the case where
+ * the users created their own journals by mounting the metafs and
+ * laying it out. But it's still possible. These journals might have
+ * several extents.
+ *
+ * TODO: This should be done in bigger chunks rather than one block at a time,
+ * but since it's only done at mount time, I'm not worried about the
+ * time it takes.
+ */
+static int map_journal_extents(struct gfs2_sbd *sdp)
+{
+ struct gfs2_jdesc *jd = sdp->sd_jdesc;
+ unsigned int lb;
+ u64 db, prev_db; /* logical block, disk block, prev disk block */
+ struct gfs2_inode *ip = GFS2_I(jd->jd_inode);
+ struct gfs2_journal_extent *jext = NULL;
+ struct buffer_head bh;
+ int rc = 0;
+
+ prev_db = 0;
+
+ for (lb = 0; lb < ip->i_di.di_size >> sdp->sd_sb.sb_bsize_shift; lb++) {
+ bh.b_state = 0;
+ bh.b_blocknr = 0;
+ bh.b_size = 1 << ip->i_inode.i_blkbits;
+ rc = gfs2_block_map(jd->jd_inode, lb, &bh, 0);
+ db = bh.b_blocknr;
+ if (rc || !db) {
+ printk(KERN_INFO "GFS2 journal mapping error %d: lb="
+ "%u db=%llu\n", rc, lb, (unsigned long long)db);
+ break;
+ }
+ if (!prev_db || db != prev_db + 1) {
+ jext = kzalloc(sizeof(struct gfs2_journal_extent),
+ GFP_KERNEL);
+ if (!jext) {
+ printk(KERN_INFO "GFS2 error: out of memory "
+ "mapping journal extents.\n");
+ rc = -ENOMEM;
+ break;
+ }
+ jext->dblock = db;
+ jext->lblock = lb;
+ jext->blocks = 1;
+ list_add_tail(&jext->extent_list, &jd->extent_list);
+ } else {
+ jext->blocks++;
+ }
+ prev_db = db;
+ }
+ return rc;
}
static int init_journal(struct gfs2_sbd *sdp, int undo)
{
struct gfs2_holder ji_gh;
struct task_struct *p;
+ struct gfs2_inode *ip;
int jindex = 1;
int error = 0;
goto fail_recoverd;
}
- error = gfs2_lookup_simple(sdp->sd_master_dir, "jindex",
- &sdp->sd_jindex);
- if (error) {
+ sdp->sd_jindex = gfs2_lookup_simple(sdp->sd_master_dir, "jindex");
+ if (IS_ERR(sdp->sd_jindex)) {
fs_err(sdp, "can't lookup journal index: %d\n", error);
- return error;
+ return PTR_ERR(sdp->sd_jindex);
}
- set_bit(GLF_STICKY, &get_v2ip(sdp->sd_jindex)->i_gl->gl_flags);
+ ip = GFS2_I(sdp->sd_jindex);
+ set_bit(GLF_STICKY, &ip->i_gl->gl_flags);
/* Load in the journal index special file */
error = -EINVAL;
if (!gfs2_jindex_size(sdp)) {
fs_err(sdp, "no journals!\n");
- goto fail_jindex;
+ goto fail_jindex;
}
if (sdp->sd_args.ar_spectator) {
sdp->sd_jdesc = gfs2_jdesc_find(sdp, 0);
- sdp->sd_log_blks_free = sdp->sd_jdesc->jd_blocks;
+ atomic_set(&sdp->sd_log_blks_free, sdp->sd_jdesc->jd_blocks);
} else {
if (sdp->sd_lockstruct.ls_jid >= gfs2_jindex_size(sdp)) {
fs_err(sdp, "can't mount journal #%u\n",
}
sdp->sd_jdesc = gfs2_jdesc_find(sdp, sdp->sd_lockstruct.ls_jid);
- error = gfs2_glock_nq_num(sdp,
- sdp->sd_lockstruct.ls_jid,
+ error = gfs2_glock_nq_num(sdp, sdp->sd_lockstruct.ls_jid,
&gfs2_journal_glops,
LM_ST_EXCLUSIVE, LM_FLAG_NOEXP,
&sdp->sd_journal_gh);
goto fail_jindex;
}
- error = gfs2_glock_nq_init(get_v2ip(sdp->sd_jdesc->jd_inode)->i_gl,
- LM_ST_SHARED,
- LM_FLAG_NOEXP | GL_EXACT,
+ ip = GFS2_I(sdp->sd_jdesc->jd_inode);
+ error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED,
+ LM_FLAG_NOEXP | GL_EXACT | GL_NOCACHE,
&sdp->sd_jinode_gh);
if (error) {
fs_err(sdp, "can't acquire journal inode glock: %d\n",
sdp->sd_jdesc->jd_jid, error);
goto fail_jinode_gh;
}
- sdp->sd_log_blks_free = sdp->sd_jdesc->jd_blocks;
+ atomic_set(&sdp->sd_log_blks_free, sdp->sd_jdesc->jd_blocks);
+
+ /* Map the extents for this journal's blocks */
+ map_journal_extents(sdp);
}
if (sdp->sd_lockstruct.ls_first) {
unsigned int x;
for (x = 0; x < sdp->sd_journals; x++) {
- error = gfs2_recover_journal(gfs2_jdesc_find(sdp, x),
- WAIT);
+ error = gfs2_recover_journal(gfs2_jdesc_find(sdp, x));
if (error) {
fs_err(sdp, "error recovering journal %u: %d\n",
x, error);
gfs2_lm_others_may_mount(sdp);
} else if (!sdp->sd_args.ar_spectator) {
- error = gfs2_recover_journal(sdp->sd_jdesc, WAIT);
+ error = gfs2_recover_journal(sdp->sd_jdesc);
if (error) {
fs_err(sdp, "error recovering my journal: %d\n", error);
goto fail_jinode_gh;
gfs2_glock_dq_uninit(&ji_gh);
jindex = 0;
- /* Disown my Journal glock */
-
- sdp->sd_journal_gh.gh_owner = NULL;
- sdp->sd_jinode_gh.gh_owner = NULL;
-
p = kthread_run(gfs2_recoverd, sdp, "gfs2_recoverd");
error = IS_ERR(p);
if (error) {
return 0;
- fail_recoverd:
+fail_recoverd:
kthread_stop(sdp->sd_recoverd_process);
-
- fail_jinode_gh:
+fail_jinode_gh:
if (!sdp->sd_args.ar_spectator)
gfs2_glock_dq_uninit(&sdp->sd_jinode_gh);
-
- fail_journal_gh:
+fail_journal_gh:
if (!sdp->sd_args.ar_spectator)
gfs2_glock_dq_uninit(&sdp->sd_journal_gh);
-
- fail_jindex:
+fail_jindex:
gfs2_jindex_free(sdp);
if (jindex)
gfs2_glock_dq_uninit(&ji_gh);
-
- fail:
+fail:
iput(sdp->sd_jindex);
-
return error;
}
static int init_inodes(struct gfs2_sbd *sdp, int undo)
{
int error = 0;
+ struct gfs2_inode *ip;
+ struct inode *inode;
if (undo)
goto fail_qinode;
- error = gfs2_lookup_master_dir(sdp);
- if (error) {
+ inode = gfs2_lookup_root(sdp->sd_vfs, sdp->sd_sb.sb_master_dir.no_addr);
+ if (IS_ERR(inode)) {
+ error = PTR_ERR(inode);
fs_err(sdp, "can't read in master directory: %d\n", error);
goto fail;
}
+ sdp->sd_master_dir = inode;
error = init_journal(sdp, undo);
if (error)
goto fail_master;
/* Read in the master inode number inode */
- error = gfs2_lookup_simple(sdp->sd_master_dir, "inum",
- &sdp->sd_inum_inode);
- if (error) {
+ sdp->sd_inum_inode = gfs2_lookup_simple(sdp->sd_master_dir, "inum");
+ if (IS_ERR(sdp->sd_inum_inode)) {
+ error = PTR_ERR(sdp->sd_inum_inode);
fs_err(sdp, "can't read in inum inode: %d\n", error);
goto fail_journal;
}
/* Read in the master statfs inode */
- error = gfs2_lookup_simple(sdp->sd_master_dir, "statfs",
- &sdp->sd_statfs_inode);
- if (error) {
+ sdp->sd_statfs_inode = gfs2_lookup_simple(sdp->sd_master_dir, "statfs");
+ if (IS_ERR(sdp->sd_statfs_inode)) {
+ error = PTR_ERR(sdp->sd_statfs_inode);
fs_err(sdp, "can't read in statfs inode: %d\n", error);
goto fail_inum;
}
/* Read in the resource index inode */
- error = gfs2_lookup_simple(sdp->sd_master_dir, "rindex",
- &sdp->sd_rindex);
- if (error) {
+ sdp->sd_rindex = gfs2_lookup_simple(sdp->sd_master_dir, "rindex");
+ if (IS_ERR(sdp->sd_rindex)) {
+ error = PTR_ERR(sdp->sd_rindex);
fs_err(sdp, "can't get resource index inode: %d\n", error);
goto fail_statfs;
}
- set_bit(GLF_STICKY, &get_v2ip(sdp->sd_rindex)->i_gl->gl_flags);
- sdp->sd_rindex_vn = get_v2ip(sdp->sd_rindex)->i_gl->gl_vn - 1;
+ ip = GFS2_I(sdp->sd_rindex);
+ set_bit(GLF_STICKY, &ip->i_gl->gl_flags);
+ sdp->sd_rindex_vn = ip->i_gl->gl_vn - 1;
/* Read in the quota inode */
- error = gfs2_lookup_simple(sdp->sd_master_dir, "quota",
- &sdp->sd_quota_inode);
- if (error) {
+ sdp->sd_quota_inode = gfs2_lookup_simple(sdp->sd_master_dir, "quota");
+ if (IS_ERR(sdp->sd_quota_inode)) {
+ error = PTR_ERR(sdp->sd_quota_inode);
fs_err(sdp, "can't get quota file inode: %d\n", error);
goto fail_rindex;
}
fail_qinode:
iput(sdp->sd_quota_inode);
-
fail_rindex:
gfs2_clear_rgrpd(sdp);
iput(sdp->sd_rindex);
-
fail_statfs:
iput(sdp->sd_statfs_inode);
-
fail_inum:
iput(sdp->sd_inum_inode);
fail_journal:
struct inode *pn = NULL;
char buf[30];
int error = 0;
+ struct gfs2_inode *ip;
if (sdp->sd_args.ar_spectator)
return 0;
if (undo)
goto fail_qc_gh;
- error = gfs2_lookup_simple(sdp->sd_master_dir, "per_node", &pn);
- if (error) {
+ pn = gfs2_lookup_simple(sdp->sd_master_dir, "per_node");
+ if (IS_ERR(pn)) {
+ error = PTR_ERR(pn);
fs_err(sdp, "can't find per_node directory: %d\n", error);
return error;
}
sprintf(buf, "inum_range%u", sdp->sd_jdesc->jd_jid);
- error = gfs2_lookup_simple(pn, buf, &sdp->sd_ir_inode);
- if (error) {
+ sdp->sd_ir_inode = gfs2_lookup_simple(pn, buf);
+ if (IS_ERR(sdp->sd_ir_inode)) {
+ error = PTR_ERR(sdp->sd_ir_inode);
fs_err(sdp, "can't find local \"ir\" file: %d\n", error);
goto fail;
}
sprintf(buf, "statfs_change%u", sdp->sd_jdesc->jd_jid);
- error = gfs2_lookup_simple(pn, buf, &sdp->sd_sc_inode);
- if (error) {
+ sdp->sd_sc_inode = gfs2_lookup_simple(pn, buf);
+ if (IS_ERR(sdp->sd_sc_inode)) {
+ error = PTR_ERR(sdp->sd_sc_inode);
fs_err(sdp, "can't find local \"sc\" file: %d\n", error);
goto fail_ir_i;
}
- sprintf(buf, "unlinked_tag%u", sdp->sd_jdesc->jd_jid);
- error = gfs2_lookup_simple(pn, buf, &sdp->sd_ut_inode);
- if (error) {
- fs_err(sdp, "can't find local \"ut\" file: %d\n", error);
- goto fail_sc_i;
- }
-
sprintf(buf, "quota_change%u", sdp->sd_jdesc->jd_jid);
- error = gfs2_lookup_simple(pn, buf, &sdp->sd_qc_inode);
- if (error) {
+ sdp->sd_qc_inode = gfs2_lookup_simple(pn, buf);
+ if (IS_ERR(sdp->sd_qc_inode)) {
+ error = PTR_ERR(sdp->sd_qc_inode);
fs_err(sdp, "can't find local \"qc\" file: %d\n", error);
goto fail_ut_i;
}
iput(pn);
pn = NULL;
- error = gfs2_glock_nq_init(get_v2ip(sdp->sd_ir_inode)->i_gl,
- LM_ST_EXCLUSIVE, GL_NEVER_RECURSE,
+ ip = GFS2_I(sdp->sd_ir_inode);
+ error = gfs2_glock_nq_init(ip->i_gl,
+ LM_ST_EXCLUSIVE, 0,
&sdp->sd_ir_gh);
if (error) {
fs_err(sdp, "can't lock local \"ir\" file: %d\n", error);
goto fail_qc_i;
}
- error = gfs2_glock_nq_init(get_v2ip(sdp->sd_sc_inode)->i_gl,
- LM_ST_EXCLUSIVE, GL_NEVER_RECURSE,
+ ip = GFS2_I(sdp->sd_sc_inode);
+ error = gfs2_glock_nq_init(ip->i_gl,
+ LM_ST_EXCLUSIVE, 0,
&sdp->sd_sc_gh);
if (error) {
fs_err(sdp, "can't lock local \"sc\" file: %d\n", error);
goto fail_ir_gh;
}
- error = gfs2_glock_nq_init(get_v2ip(sdp->sd_ut_inode)->i_gl,
- LM_ST_EXCLUSIVE, GL_NEVER_RECURSE,
- &sdp->sd_ut_gh);
- if (error) {
- fs_err(sdp, "can't lock local \"ut\" file: %d\n", error);
- goto fail_sc_gh;
- }
-
- error = gfs2_glock_nq_init(get_v2ip(sdp->sd_qc_inode)->i_gl,
- LM_ST_EXCLUSIVE, GL_NEVER_RECURSE,
+ ip = GFS2_I(sdp->sd_qc_inode);
+ error = gfs2_glock_nq_init(ip->i_gl,
+ LM_ST_EXCLUSIVE, 0,
&sdp->sd_qc_gh);
if (error) {
fs_err(sdp, "can't lock local \"qc\" file: %d\n", error);
return 0;
- fail_qc_gh:
+fail_qc_gh:
gfs2_glock_dq_uninit(&sdp->sd_qc_gh);
-
- fail_ut_gh:
- gfs2_glock_dq_uninit(&sdp->sd_ut_gh);
-
- fail_sc_gh:
+fail_ut_gh:
gfs2_glock_dq_uninit(&sdp->sd_sc_gh);
-
- fail_ir_gh:
+fail_ir_gh:
gfs2_glock_dq_uninit(&sdp->sd_ir_gh);
-
- fail_qc_i:
+fail_qc_i:
iput(sdp->sd_qc_inode);
-
- fail_ut_i:
- iput(sdp->sd_ut_inode);
-
- fail_sc_i:
+fail_ut_i:
iput(sdp->sd_sc_inode);
-
- fail_ir_i:
+fail_ir_i:
iput(sdp->sd_ir_inode);
-
- fail:
+fail:
if (pn)
iput(pn);
return error;
int error = 0;
if (undo)
- goto fail_inoded;
+ goto fail_quotad;
sdp->sd_log_flush_time = jiffies;
sdp->sd_jindex_refresh_time = jiffies;
}
sdp->sd_quotad_process = p;
- p = kthread_run(gfs2_inoded, sdp, "gfs2_inoded");
- error = IS_ERR(p);
- if (error) {
- fs_err(sdp, "can't start inoded thread: %d\n", error);
- goto fail_quotad;
- }
- sdp->sd_inoded_process = p;
-
return 0;
- fail_inoded:
- kthread_stop(sdp->sd_inoded_process);
- fail_quotad:
+fail_quotad:
kthread_stop(sdp->sd_quotad_process);
-
- fail:
+fail:
kthread_stop(sdp->sd_logd_process);
-
return error;
}
goto fail;
}
- init_vfs(sdp);
+ init_vfs(sb, SDF_NOATIME);
+
+ /* Set up the buffer cache and fill in some fake block size values
+ to allow us to read-in the on-disk superblock. */
+ sdp->sd_sb.sb_bsize = sb_min_blocksize(sb, GFS2_BASIC_BLOCK);
+ sdp->sd_sb.sb_bsize_shift = sb->s_blocksize_bits;
+ sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift -
+ GFS2_BASIC_BLOCK_SHIFT;
+ sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift;
error = init_names(sdp, silent);
if (error)
goto fail;
+ gfs2_create_debugfs_file(sdp);
+
error = gfs2_sys_fs_add(sdp);
if (error)
goto fail;
return 0;
- fail_threads:
+fail_threads:
init_threads(sdp, UNDO);
-
- fail_per_node:
+fail_per_node:
init_per_node(sdp, UNDO);
-
- fail_inodes:
+fail_inodes:
init_inodes(sdp, UNDO);
-
- fail_sb:
+fail_sb:
init_sb(sdp, 0, UNDO);
-
- fail_locking:
+fail_locking:
init_locking(sdp, &mount_gh, UNDO);
-
- fail_lm:
+fail_lm:
gfs2_gl_hash_clear(sdp, WAIT);
gfs2_lm_unmount(sdp);
while (invalidate_inodes(sb))
yield();
-
- fail_sys:
+fail_sys:
gfs2_sys_fs_del(sdp);
+fail:
+ gfs2_delete_debugfs_file(sdp);
+ kfree(sdp);
+ sb->s_fs_info = NULL;
+ return error;
+}
+
+static int gfs2_get_sb(struct file_system_type *fs_type, int flags,
+ const char *dev_name, void *data, struct vfsmount *mnt)
+{
+ struct super_block *sb;
+ struct gfs2_sbd *sdp;
+ int error = get_sb_bdev(fs_type, flags, dev_name, data, fill_super, mnt);
+ if (error)
+ goto out;
+ sb = mnt->mnt_sb;
+ sdp = sb->s_fs_info;
+ sdp->sd_gfs2mnt = mnt;
+out:
+ return error;
+}
+
+static int fill_super_meta(struct super_block *sb, struct super_block *new,
+ void *data, int silent)
+{
+ struct gfs2_sbd *sdp = sb->s_fs_info;
+ struct inode *inode;
+ int error = 0;
- fail:
- vfree(sdp);
- set_v2sdp(sb, NULL);
+ new->s_fs_info = sdp;
+ sdp->sd_vfs_meta = sb;
+
+ init_vfs(new, SDF_NOATIME);
+
+ /* Get the master inode */
+ inode = igrab(sdp->sd_master_dir);
+
+ new->s_root = d_alloc_root(inode);
+ if (!new->s_root) {
+ fs_err(sdp, "can't get root dentry\n");
+ error = -ENOMEM;
+ iput(inode);
+ } else
+ new->s_root->d_op = &gfs2_dops;
return error;
}
-static struct super_block *gfs2_get_sb(struct file_system_type *fs_type,
- int flags, const char *dev_name,
- void *data)
+static int set_bdev_super(struct super_block *s, void *data)
+{
+ s->s_bdev = data;
+ s->s_dev = s->s_bdev->bd_dev;
+ return 0;
+}
+
+static int test_bdev_super(struct super_block *s, void *data)
+{
+ return s->s_bdev == data;
+}
+
+static struct super_block* get_gfs2_sb(const char *dev_name)
{
- return get_sb_bdev(fs_type, flags, dev_name, data, fill_super);
+ struct kstat stat;
+ struct nameidata nd;
+ struct file_system_type *fstype;
+ struct super_block *sb = NULL, *s;
+ int error;
+
+ error = path_lookup(dev_name, LOOKUP_FOLLOW, &nd);
+ if (error) {
+ printk(KERN_WARNING "GFS2: path_lookup on %s returned error\n",
+ dev_name);
+ goto out;
+ }
+ error = vfs_getattr(nd.mnt, nd.dentry, &stat);
+
+ fstype = get_fs_type("gfs2");
+ list_for_each_entry(s, &fstype->fs_supers, s_instances) {
+ if ((S_ISBLK(stat.mode) && s->s_dev == stat.rdev) ||
+ (S_ISDIR(stat.mode) && s == nd.dentry->d_inode->i_sb)) {
+ sb = s;
+ goto free_nd;
+ }
+ }
+
+ printk(KERN_WARNING "GFS2: Unrecognized block device or "
+ "mount point %s\n", dev_name);
+
+free_nd:
+ path_release(&nd);
+out:
+ return sb;
+}
+
+static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags,
+ const char *dev_name, void *data, struct vfsmount *mnt)
+{
+ int error = 0;
+ struct super_block *sb = NULL, *new;
+ struct gfs2_sbd *sdp;
+
+ sb = get_gfs2_sb(dev_name);
+ if (!sb) {
+ printk(KERN_WARNING "GFS2: gfs2 mount does not exist\n");
+ error = -ENOENT;
+ goto error;
+ }
+ sdp = sb->s_fs_info;
+ if (sdp->sd_vfs_meta) {
+ printk(KERN_WARNING "GFS2: gfs2meta mount already exists\n");
+ error = -EBUSY;
+ goto error;
+ }
+ down(&sb->s_bdev->bd_mount_sem);
+ new = sget(fs_type, test_bdev_super, set_bdev_super, sb->s_bdev);
+ up(&sb->s_bdev->bd_mount_sem);
+ if (IS_ERR(new)) {
+ error = PTR_ERR(new);
+ goto error;
+ }
+ module_put(fs_type->owner);
+ new->s_flags = flags;
+ strlcpy(new->s_id, sb->s_id, sizeof(new->s_id));
+ sb_set_blocksize(new, sb->s_blocksize);
+ error = fill_super_meta(sb, new, data, flags & MS_SILENT ? 1 : 0);
+ if (error) {
+ up_write(&new->s_umount);
+ deactivate_super(new);
+ goto error;
+ }
+
+ new->s_flags |= MS_ACTIVE;
+
+ /* Grab a reference to the gfs2 mount point */
+ atomic_inc(&sdp->sd_gfs2mnt->mnt_count);
+ return simple_set_mnt(mnt, new);
+error:
+ return error;
+}
+
+static void gfs2_kill_sb(struct super_block *sb)
+{
+ if (sb->s_fs_info) {
+ gfs2_delete_debugfs_file(sb->s_fs_info);
+ gfs2_meta_syncfs(sb->s_fs_info);
+ }
+ kill_block_super(sb);
+}
+
+static void gfs2_kill_sb_meta(struct super_block *sb)
+{
+ struct gfs2_sbd *sdp = sb->s_fs_info;
+ generic_shutdown_super(sb);
+ sdp->sd_vfs_meta = NULL;
+ atomic_dec(&sdp->sd_gfs2mnt->mnt_count);
}
struct file_system_type gfs2_fs_type = {
.name = "gfs2",
.fs_flags = FS_REQUIRES_DEV,
.get_sb = gfs2_get_sb,
- .kill_sb = kill_block_super,
+ .kill_sb = gfs2_kill_sb,
+ .owner = THIS_MODULE,
+};
+
+struct file_system_type gfs2meta_fs_type = {
+ .name = "gfs2meta",
+ .fs_flags = FS_REQUIRES_DEV,
+ .get_sb = gfs2_get_sb_meta,
+ .kill_sb = gfs2_kill_sb_meta,
.owner = THIS_MODULE,
};