GFS2: update freeze code to use freeze/thaw_super on all nodes

The current gfs2 freezing code is considerably more complicated than it
should be because it doesn't use the vfs freezing code on any node except
the one that begins the freeze.  This is because it needs to acquire a
cluster glock before calling the vfs code to prevent a deadlock, and
without the new freeze_super and thaw_super hooks, that was impossible. To
deal with the issue, gfs2 had to do some hacky locking tricks to make sure
that a frozen node couldn't be holding on a lock it needed to do the
unfreeze ioctl.

This patch makes use of the new hooks to simply the gfs2 locking code. Now,
all the nodes in the cluster freeze and thaw in exactly the same way. Every
node in the cluster caches the freeze glock in the shared state.  The new
freeze_super hook allows the freezing node to grab this freeze glock in
the exclusive state without first calling the vfs freeze_super function.
All the nodes in the cluster see this lock change, and call the vfs
freeze_super function. The vfs locking code guarantees that the nodes can't
get stuck holding the glocks necessary to unfreeze the system.  To
unfreeze, the freezing node uses the new thaw_super hook to drop the freeze
glock. Again, all the nodes notice this, reacquire the glock in shared mode
and call the vfs thaw_super function.

Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

10 files changed, 161 insertions, 126 deletions

diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 1cc0bba..fe91951 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -28,6 +28,8 @@
 #include "trans.h"
 #include "dir.h"
 
+struct workqueue_struct *gfs2_freeze_wq;
+
 static void gfs2_ail_error(struct gfs2_glock *gl, const struct buffer_head *bh)
 {
 	fs_err(gl->gl_sbd, "AIL buffer %p: blocknr %llu state 0x%08lx mapping %p page state 0x%lx\n",
@@ -94,11 +96,8 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
          * on the stack */
 	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
 	tr.tr_ip = _RET_IP_;
-	sb_start_intwrite(sdp->sd_vfs);
-	if (gfs2_log_reserve(sdp, tr.tr_reserved) < 0) {
-		sb_end_intwrite(sdp->sd_vfs);
+	if (gfs2_log_reserve(sdp, tr.tr_reserved) < 0)
 		return;
-	}
 	WARN_ON_ONCE(current->journal_info);
 	current->journal_info = &tr;
 
@@ -469,20 +468,19 @@ static void inode_go_dump(struct seq_file *seq, const struct gfs2_glock *gl)
 
 static void freeze_go_sync(struct gfs2_glock *gl)
 {
+	int error = 0;
 	struct gfs2_sbd *sdp = gl->gl_sbd;
-	DEFINE_WAIT(wait);
 
 	if (gl->gl_state == LM_ST_SHARED &&
 	    test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) {
-		atomic_set(&sdp->sd_log_freeze, 1);
-		wake_up(&sdp->sd_logd_waitq);
-		do {
-			prepare_to_wait(&sdp->sd_log_frozen_wait, &wait,
-					TASK_UNINTERRUPTIBLE);
-			if (atomic_read(&sdp->sd_log_freeze))
-				io_schedule();
-		} while(atomic_read(&sdp->sd_log_freeze));
-		finish_wait(&sdp->sd_log_frozen_wait, &wait);
+		atomic_set(&sdp->sd_freeze_state, SFS_STARTING_FREEZE);
+		error = freeze_super(sdp->sd_vfs);
+		if (error) {
+			printk(KERN_INFO "GFS2: couldn't freeze filesystem: %d\n", error);
+			gfs2_assert_withdraw(sdp, 0);
+		}
+		queue_work(gfs2_freeze_wq, &sdp->sd_freeze_work);
+		gfs2_log_flush(sdp, NULL, FREEZE_FLUSH);
 	}
 }
 
diff --git a/fs/gfs2/glops.h b/fs/gfs2/glops.h
index 7455d26..8ed1857 100644
--- a/fs/gfs2/glops.h
+++ b/fs/gfs2/glops.h
@@ -12,6 +12,8 @@
 
 #include "incore.h"
 
+extern struct workqueue_struct *gfs2_freeze_wq;
+
 extern const struct gfs2_glock_operations gfs2_meta_glops;
 extern const struct gfs2_glock_operations gfs2_inode_glops;
 extern const struct gfs2_glock_operations gfs2_rgrp_glops;
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 1b89918..7a2dbbc 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -588,6 +588,12 @@ enum {
 	SDF_SKIP_DLM_UNLOCK	= 8,
 };
 
+enum gfs2_freeze_state {
+	SFS_UNFROZEN		= 0,
+	SFS_STARTING_FREEZE	= 1,
+	SFS_FROZEN		= 2,
+};
+
 #define GFS2_FSNAME_LEN		256
 
 struct gfs2_inum_host {
@@ -685,6 +691,7 @@ struct gfs2_sbd {
 	struct gfs2_holder sd_live_gh;
 	struct gfs2_glock *sd_rename_gl;
 	struct gfs2_glock *sd_freeze_gl;
+	struct work_struct sd_freeze_work;
 	wait_queue_head_t sd_glock_wait;
 	atomic_t sd_glock_disposal;
 	struct completion sd_locking_init;
@@ -789,6 +796,9 @@ struct gfs2_sbd {
 	wait_queue_head_t sd_log_flush_wait;
 	int sd_log_error;
 
+	atomic_t sd_reserving_log;
+	wait_queue_head_t sd_reserving_log_wait;
+
 	unsigned int sd_log_flush_head;
 	u64 sd_log_flush_wrapped;
 
@@ -798,12 +808,8 @@ struct gfs2_sbd {
 
 	/* For quiescing the filesystem */
 	struct gfs2_holder sd_freeze_gh;
-	struct gfs2_holder sd_freeze_root_gh;
-	struct gfs2_holder sd_thaw_gh;
-	atomic_t sd_log_freeze;
-	atomic_t sd_frozen_root;
-	wait_queue_head_t sd_frozen_root_wait;
-	wait_queue_head_t sd_log_frozen_wait;
+	atomic_t sd_freeze_state;
+	struct mutex sd_freeze_mutex;
 
 	char sd_fsname[GFS2_FSNAME_LEN];
 	char sd_table_name[GFS2_FSNAME_LEN];
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index b41b5c7..04065e5 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -1618,26 +1618,18 @@ int gfs2_permission(struct inode *inode, int mask)
 {
 	struct gfs2_inode *ip;
 	struct gfs2_holder i_gh;
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	int error;
 	int unlock = 0;
-	int frozen_root = 0;
 
 
 	ip = GFS2_I(inode);
 	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
-		if (unlikely(gfs2_glock_is_held_excl(sdp->sd_freeze_gl) &&
-			     inode == sdp->sd_root_dir->d_inode &&
-			     atomic_inc_not_zero(&sdp->sd_frozen_root)))
-			frozen_root = 1;
-		else {
-			if (mask & MAY_NOT_BLOCK)
-				return -ECHILD;
-			error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
-			if (error)
-				return error;
-			unlock = 1;
-		}
+		if (mask & MAY_NOT_BLOCK)
+			return -ECHILD;
+		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
+		if (error)
+			return error;
+		unlock = 1;
 	}
 
 	if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode))
@@ -1646,8 +1638,6 @@ int gfs2_permission(struct inode *inode, int mask)
 		error = generic_permission(inode, mask);
 	if (unlock)
 		gfs2_glock_dq_uninit(&i_gh);
-	else if (frozen_root && atomic_dec_and_test(&sdp->sd_frozen_root))
-		wake_up(&sdp->sd_frozen_root_wait);
 
 	return error;
 }
@@ -1820,29 +1810,19 @@ static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	struct inode *inode = dentry->d_inode;
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_holder gh;
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	int error;
 	int unlock = 0;
-	int frozen_root = 0;
 
 	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
-		if (unlikely(gfs2_glock_is_held_excl(sdp->sd_freeze_gl) &&
-			     inode == sdp->sd_root_dir->d_inode &&
-			     atomic_inc_not_zero(&sdp->sd_frozen_root)))
-			frozen_root = 1;
-		else {
-			error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
-			if (error)
-				return error;
-			unlock = 1;
-		}
+		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
+		if (error)
+			return error;
+		unlock = 1;
 	}
 
 	generic_fillattr(inode, stat);
 	if (unlock)
 		gfs2_glock_dq_uninit(&gh);
-	else if (frozen_root && atomic_dec_and_test(&sdp->sd_frozen_root))
-		wake_up(&sdp->sd_frozen_root_wait);
 
 	return 0;
 }
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 3966fad..536e7a6 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -339,6 +339,7 @@ void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks)
 
 int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks)
 {
+	int ret = 0;
 	unsigned reserved_blks = 7 * (4096 / sdp->sd_vfs->s_blocksize);
 	unsigned wanted = blks + reserved_blks;
 	DEFINE_WAIT(wait);
@@ -362,9 +363,13 @@ retry:
 		} while(free_blocks <= wanted);
 		finish_wait(&sdp->sd_log_waitq, &wait);
 	}
+	atomic_inc(&sdp->sd_reserving_log);
 	if (atomic_cmpxchg(&sdp->sd_log_blks_free, free_blocks,
-				free_blocks - blks) != free_blocks)
+				free_blocks - blks) != free_blocks) {
+		if (atomic_dec_and_test(&sdp->sd_reserving_log))
+			wake_up(&sdp->sd_reserving_log_wait);
 		goto retry;
+	}
 	trace_gfs2_log_blocks(sdp, -blks);
 
 	/*
@@ -377,9 +382,11 @@ retry:
 	down_read(&sdp->sd_log_flush_lock);
 	if (unlikely(!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))) {
 		gfs2_log_release(sdp, blks);
-		return -EROFS;
+		ret = -EROFS;
 	}
-	return 0;
+	if (atomic_dec_and_test(&sdp->sd_reserving_log))
+		wake_up(&sdp->sd_reserving_log_wait);
+	return ret;
 }
 
 /**
@@ -652,9 +659,12 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags)
 	u32 hash;
 	int rw = WRITE_FLUSH_FUA | REQ_META;
 	struct page *page = mempool_alloc(gfs2_page_pool, GFP_NOIO);
+	enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state);
 	lh = page_address(page);
 	clear_page(lh);
 
+	gfs2_assert_withdraw(sdp, (state != SFS_FROZEN));
+
 	tail = current_tail(sdp);
 
 	lh->lh_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
@@ -695,6 +705,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 		    enum gfs2_flush_type type)
 {
 	struct gfs2_trans *tr;
+	enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state);
 
 	down_write(&sdp->sd_log_flush_lock);
 
@@ -713,8 +724,12 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 		INIT_LIST_HEAD(&tr->tr_ail1_list);
 		INIT_LIST_HEAD(&tr->tr_ail2_list);
 		tr->tr_first = sdp->sd_log_flush_head;
+		if (unlikely (state == SFS_FROZEN))
+			gfs2_assert_withdraw(sdp, !tr->tr_num_buf_new && !tr->tr_num_databuf_new);
 	}
 
+	if (unlikely(state == SFS_FROZEN))
+		gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
 	gfs2_assert_withdraw(sdp,
 			sdp->sd_log_num_revoke == sdp->sd_log_commited_revoke);
 
@@ -745,8 +760,6 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 	spin_unlock(&sdp->sd_ail_lock);
 	gfs2_log_unlock(sdp);
 
-	if (atomic_read(&sdp->sd_log_freeze))
-		type = FREEZE_FLUSH;
 	if (type != NORMAL_FLUSH) {
 		if (!sdp->sd_log_idle) {
 			for (;;) {
@@ -763,21 +776,8 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl,
 		}
 		if (type == SHUTDOWN_FLUSH || type == FREEZE_FLUSH)
 			gfs2_log_shutdown(sdp);
-		if (type == FREEZE_FLUSH) {
-			int error;
-
-			atomic_set(&sdp->sd_log_freeze, 0);
-			wake_up(&sdp->sd_log_frozen_wait);
-			error = gfs2_glock_nq_init(sdp->sd_freeze_gl,
-						   LM_ST_SHARED, 0,
-						   &sdp->sd_thaw_gh);
-			if (error) {
-				printk(KERN_INFO "GFS2: couln't get freeze lock : %d\n", error);
-				gfs2_assert_withdraw(sdp, 0);
-			}
-			else
-				gfs2_glock_dq_uninit(&sdp->sd_thaw_gh);
-		}
+		if (type == FREEZE_FLUSH)
+			atomic_set(&sdp->sd_freeze_state, SFS_FROZEN);
 	}
 
 	trace_gfs2_log_flush(sdp, 0);
@@ -888,7 +888,7 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp)
 
 static inline int gfs2_jrnl_flush_reqd(struct gfs2_sbd *sdp)
 {
-	return (atomic_read(&sdp->sd_log_pinned) >= atomic_read(&sdp->sd_log_thresh1) || atomic_read(&sdp->sd_log_freeze));
+	return (atomic_read(&sdp->sd_log_pinned) >= atomic_read(&sdp->sd_log_thresh1));
 }
 
 static inline int gfs2_ail_flush_reqd(struct gfs2_sbd *sdp)
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 82b6ac8..241a399 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -30,6 +30,7 @@
 #include "quota.h"
 #include "recovery.h"
 #include "dir.h"
+#include "glops.h"
 
 struct workqueue_struct *gfs2_control_wq;
 
@@ -161,9 +162,14 @@ static int __init init_gfs2_fs(void)
 	if (!gfs2_control_wq)
 		goto fail_recovery;
 
+	gfs2_freeze_wq = alloc_workqueue("freeze_workqueue", 0, 0);
+
+	if (!gfs2_freeze_wq)
+		goto fail_control;
+
 	gfs2_page_pool = mempool_create_page_pool(64, 0);
 	if (!gfs2_page_pool)
-		goto fail_control;
+		goto fail_freeze;
 
 	gfs2_register_debugfs();
 
@@ -171,6 +177,8 @@ static int __init init_gfs2_fs(void)
 
 	return 0;
 
+fail_freeze:
+	destroy_workqueue(gfs2_freeze_wq);
 fail_control:
 	destroy_workqueue(gfs2_control_wq);
 fail_recovery:
@@ -224,6 +232,7 @@ static void __exit exit_gfs2_fs(void)
 	unregister_filesystem(&gfs2meta_fs_type);
 	destroy_workqueue(gfs_recovery_wq);
 	destroy_workqueue(gfs2_control_wq);
+	destroy_workqueue(gfs2_freeze_wq);
 	list_lru_destroy(&gfs2_qd_lru);
 
 	rcu_barrier();
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index d3eae24..b5803ac 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -129,11 +129,11 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
 
 	init_rwsem(&sdp->sd_log_flush_lock);
 	atomic_set(&sdp->sd_log_in_flight, 0);
+	atomic_set(&sdp->sd_reserving_log, 0);
+	init_waitqueue_head(&sdp->sd_reserving_log_wait);
 	init_waitqueue_head(&sdp->sd_log_flush_wait);
-	init_waitqueue_head(&sdp->sd_log_frozen_wait);
-	atomic_set(&sdp->sd_log_freeze, 0);
-	atomic_set(&sdp->sd_frozen_root, 0);
-	init_waitqueue_head(&sdp->sd_frozen_root_wait);
+	atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN);
+	mutex_init(&sdp->sd_freeze_mutex);
 
 	return sdp;
 }
@@ -760,15 +760,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo)
 	set_bit(SDF_JOURNAL_CHECKED, &sdp->sd_flags);
 	gfs2_glock_dq_uninit(&ji_gh);
 	jindex = 0;
-	if (!sdp->sd_args.ar_spectator) {
-		error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, 0,
-					   &sdp->sd_thaw_gh);
-		if (error) {
-			fs_err(sdp, "can't acquire freeze glock: %d\n", error);
-			goto fail_jinode_gh;
-		}
-	}
-	gfs2_glock_dq_uninit(&sdp->sd_thaw_gh);
+	INIT_WORK(&sdp->sd_freeze_work, gfs2_freeze_func);
 	return 0;
 
 fail_jinode_gh:
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index a346f56..5b327f8 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -26,6 +26,7 @@
 #include <linux/wait.h>
 #include <linux/writeback.h>
 #include <linux/backing-dev.h>
+#include <linux/kernel.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -399,7 +400,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 {
 	struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode);
 	struct gfs2_glock *j_gl = ip->i_gl;
-	struct gfs2_holder thaw_gh;
+	struct gfs2_holder freeze_gh;
 	struct gfs2_log_header_host head;
 	int error;
 
@@ -408,7 +409,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 		return error;
 
 	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, 0,
-				   &thaw_gh);
+				   &freeze_gh);
 	if (error)
 		goto fail_threads;
 
@@ -434,13 +435,13 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp)
 
 	set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
 
-	gfs2_glock_dq_uninit(&thaw_gh);
+	gfs2_glock_dq_uninit(&freeze_gh);
 
 	return 0;
 
 fail:
-	thaw_gh.gh_flags |= GL_NOCACHE;
-	gfs2_glock_dq_uninit(&thaw_gh);
+	freeze_gh.gh_flags |= GL_NOCACHE;
+	gfs2_glock_dq_uninit(&freeze_gh);
 fail_threads:
 	kthread_stop(sdp->sd_quotad_process);
 	kthread_stop(sdp->sd_logd_process);
@@ -580,14 +581,15 @@ int gfs2_statfs_sync(struct super_block *sb, int type)
 	struct buffer_head *m_bh, *l_bh;
 	int error;
 
+	sb_start_write(sb);
 	error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, GL_NOCACHE,
 				   &gh);
 	if (error)
-		return error;
+		goto out;
 
 	error = gfs2_meta_inode_buffer(m_ip, &m_bh);
 	if (error)
-		goto out;
+		goto out_unlock;
 
 	spin_lock(&sdp->sd_statfs_spin);
 	gfs2_statfs_change_in(m_sc, m_bh->b_data +
@@ -615,8 +617,10 @@ out_bh2:
 	brelse(l_bh);
 out_bh:
 	brelse(m_bh);
-out:
+out_unlock:
 	gfs2_glock_dq_uninit(&gh);
+out:
+	sb_end_write(sb);
 	return error;
 }
 
@@ -643,14 +647,8 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp,
 	struct lfcc *lfcc;
 	LIST_HEAD(list);
 	struct gfs2_log_header_host lh;
-	struct gfs2_inode *dip = GFS2_I(sdp->sd_root_dir->d_inode);
 	int error;
 
-	error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0,
-				   &sdp->sd_freeze_root_gh);
-	if (error)
-		return error;
-	atomic_set(&sdp->sd_frozen_root, 1);
 	list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) {
 		lfcc = kmalloc(sizeof(struct lfcc), GFP_KERNEL);
 		if (!lfcc) {
@@ -692,11 +690,6 @@ out:
 		gfs2_glock_dq_uninit(&lfcc->gh);
 		kfree(lfcc);
 	}
-	if (error) {
-		atomic_dec(&sdp->sd_frozen_root);
-		wait_event(sdp->sd_frozen_root_wait, atomic_read(&sdp->sd_frozen_root) == 0);
-		gfs2_glock_dq_uninit(&sdp->sd_freeze_root_gh);
-	}
 	return error;
 }
 
@@ -834,18 +827,14 @@ out:
 
 static int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
 {
-	struct gfs2_holder thaw_gh;
+	struct gfs2_holder freeze_gh;
 	int error;
 
 	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, GL_NOCACHE,
-				   &thaw_gh);
+				   &freeze_gh);
 	if (error && !test_bit(SDF_SHUTDOWN, &sdp->sd_flags))
 		return error;
 
-	down_write(&sdp->sd_log_flush_lock);
-	clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
-	up_write(&sdp->sd_log_flush_lock);
-
 	kthread_stop(sdp->sd_quotad_process);
 	kthread_stop(sdp->sd_logd_process);
 
@@ -853,11 +842,16 @@ static int gfs2_make_fs_ro(struct gfs2_sbd *sdp)
 	gfs2_quota_sync(sdp->sd_vfs, 0);
 	gfs2_statfs_sync(sdp->sd_vfs, 0);
 
+	down_write(&sdp->sd_log_flush_lock);
+	clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags);
+	up_write(&sdp->sd_log_flush_lock);
+
 	gfs2_log_flush(sdp, NULL, SHUTDOWN_FLUSH);
+	wait_event(sdp->sd_reserving_log_wait, atomic_read(&sdp->sd_reserving_log) == 0);
 	gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == sdp->sd_jdesc->jd_blocks);
 
-	if (thaw_gh.gh_gl)
-		gfs2_glock_dq_uninit(&thaw_gh);
+	if (freeze_gh.gh_gl)
+		gfs2_glock_dq_uninit(&freeze_gh);
 
 	gfs2_quota_cleanup(sdp);
 
@@ -943,11 +937,41 @@ static int gfs2_sync_fs(struct super_block *sb, int wait)
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 
 	gfs2_quota_sync(sb, -1);
-	if (wait && sdp && !atomic_read(&sdp->sd_log_freeze))
+	if (wait && sdp)
 		gfs2_log_flush(sdp, NULL, NORMAL_FLUSH);
 	return 0;
 }
 
+void gfs2_freeze_func(struct work_struct *work)
+{
+	int error;
+	struct gfs2_holder freeze_gh;
+	struct gfs2_sbd *sdp = container_of(work, struct gfs2_sbd, sd_freeze_work);
+	struct super_block *sb = sdp->sd_vfs;
+
+	atomic_inc(&sb->s_active);
+	error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, 0,
+				   &freeze_gh);
+	if (error) {
+		printk(KERN_INFO "GFS2: couln't get freeze lock : %d\n", error);
+		gfs2_assert_withdraw(sdp, 0);
+	}
+	else {
+		atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN);
+		error = thaw_super(sb);
+		if (error) {
+			printk(KERN_INFO "GFS2: couldn't thaw filesystem: %d\n",
+			       error);
+			gfs2_assert_withdraw(sdp, 0);
+		}
+		if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))
+			freeze_gh.gh_flags |= GL_NOCACHE;
+		gfs2_glock_dq_uninit(&freeze_gh);
+	}
+	deactivate_super(sb);
+	return;
+}
+
 /**
  * gfs2_freeze - prevent further writes to the filesystem
  * @sb: the VFS structure for the filesystem
@@ -957,10 +981,16 @@ static int gfs2_sync_fs(struct super_block *sb, int wait)
 static int gfs2_freeze(struct super_block *sb)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
-	int error;
+	int error = 0;
 
-	if (test_bit(SDF_SHUTDOWN, &sdp->sd_flags))
-		return -EINVAL;
+	mutex_lock(&sdp->sd_freeze_mutex);
+	if (atomic_read(&sdp->sd_freeze_state) != SFS_UNFROZEN)
+		goto out;
+
+	if (test_bit(SDF_SHUTDOWN, &sdp->sd_flags)) {
+		error = -EINVAL;
+		goto out;
+	}
 
 	for (;;) {
 		error = gfs2_lock_fs_check_clean(sdp, &sdp->sd_freeze_gh);
@@ -980,7 +1010,10 @@ static int gfs2_freeze(struct super_block *sb)
 		fs_err(sdp, "retrying...\n");
 		msleep(1000);
 	}
-	return 0;
+	error = 0;
+out:
+	mutex_unlock(&sdp->sd_freeze_mutex);
+	return error;
 }
 
 /**
@@ -993,10 +1026,15 @@ static int gfs2_unfreeze(struct super_block *sb)
 {
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 
+	mutex_lock(&sdp->sd_freeze_mutex);
+        if (atomic_read(&sdp->sd_freeze_state) != SFS_FROZEN ||
+	    sdp->sd_freeze_gh.gh_gl == NULL) {
+		mutex_unlock(&sdp->sd_freeze_mutex);
+                return 0;
+	}
+
 	gfs2_glock_dq_uninit(&sdp->sd_freeze_gh);
-	atomic_dec(&sdp->sd_frozen_root);
-	wait_event(sdp->sd_frozen_root_wait, atomic_read(&sdp->sd_frozen_root) == 0);
-	gfs2_glock_dq_uninit(&sdp->sd_freeze_root_gh);
+	mutex_unlock(&sdp->sd_freeze_mutex);
 	return 0;
 }
 
@@ -1618,8 +1656,8 @@ const struct super_operations gfs2_super_ops = {
 	.evict_inode		= gfs2_evict_inode,
 	.put_super		= gfs2_put_super,
 	.sync_fs		= gfs2_sync_fs,
-	.freeze_fs 		= gfs2_freeze,
-	.unfreeze_fs		= gfs2_unfreeze,
+	.freeze_super		= gfs2_freeze,
+	.thaw_super		= gfs2_unfreeze,
 	.statfs			= gfs2_statfs,
 	.remount_fs		= gfs2_remount_fs,
 	.drop_inode		= gfs2_drop_inode,
diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h
index 90e3322..73c97dc 100644
--- a/fs/gfs2/super.h
+++ b/fs/gfs2/super.h
@@ -45,6 +45,7 @@ extern void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc,
 extern void update_statfs(struct gfs2_sbd *sdp, struct buffer_head *m_bh,
 			  struct buffer_head *l_bh);
 extern int gfs2_statfs_sync(struct super_block *sb, int type);
+extern void gfs2_freeze_func(struct work_struct *work);
 
 extern struct file_system_type gfs2_fs_type;
 extern struct file_system_type gfs2meta_fs_type;
diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c
index 42bfd336..88bff24 100644
--- a/fs/gfs2/trans.c
+++ b/fs/gfs2/trans.c
@@ -89,14 +89,17 @@ void gfs2_trans_end(struct gfs2_sbd *sdp)
 {
 	struct gfs2_trans *tr = current->journal_info;
 	s64 nbuf;
+	int alloced = tr->tr_alloced;
+
 	BUG_ON(!tr);
 	current->journal_info = NULL;
 
 	if (!tr->tr_touched) {
 		gfs2_log_release(sdp, tr->tr_reserved);
-		if (tr->tr_alloced)
+		if (alloced) {
 			kfree(tr);
-		sb_end_intwrite(sdp->sd_vfs);
+			sb_end_intwrite(sdp->sd_vfs);
+		}
 		return;
 	}
 
@@ -109,13 +112,14 @@ void gfs2_trans_end(struct gfs2_sbd *sdp)
 		gfs2_print_trans(tr);
 
 	gfs2_log_commit(sdp, tr);
-	if (tr->tr_alloced && !tr->tr_attached)
+	if (alloced && !tr->tr_attached)
 			kfree(tr);
 	up_read(&sdp->sd_log_flush_lock);
 
 	if (sdp->sd_vfs->s_flags & MS_SYNCHRONOUS)
 		gfs2_log_flush(sdp, NULL, NORMAL_FLUSH);
-	sb_end_intwrite(sdp->sd_vfs);
+	if (alloced)
+		sb_end_intwrite(sdp->sd_vfs);
 }
 
 static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl,
@@ -192,6 +196,7 @@ static void meta_lo_add(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd)
 {
 	struct gfs2_meta_header *mh;
 	struct gfs2_trans *tr;
+	enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state);
 
 	tr = current->journal_info;
 	tr->tr_touched = 1;
@@ -205,6 +210,10 @@ static void meta_lo_add(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd)
 		       (unsigned long long)bd->bd_bh->b_blocknr);
 		BUG();
 	}
+	if (unlikely(state == SFS_FROZEN)) {
+		printk(KERN_INFO "GFS2:adding buf while frozen\n");
+		gfs2_assert_withdraw(sdp, 0);
+	}
 	gfs2_pin(sdp, bd->bd_bh);
 	mh->__pad0 = cpu_to_be64(0);
 	mh->mh_jid = cpu_to_be32(sdp->sd_jdesc->jd_jid);