Linux-2.6.12-rc2

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!

7 files changed, 6845 insertions, 0 deletions

diff --git a/fs/jbd/Makefile b/fs/jbd/Makefile
new file mode 100644
index 00000000000..54aca4868a3
--- /dev/null
+++ b/fs/jbd/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for the linux journaling routines.
+#
+
+obj-$(CONFIG_JBD) += jbd.o
+
+jbd-objs := transaction.o commit.o recovery.o checkpoint.o revoke.o journal.o
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
new file mode 100644
index 00000000000..98d830401c5
--- /dev/null
+++ b/fs/jbd/checkpoint.c
@@ -0,0 +1,636 @@
+/*
+ * linux/fs/checkpoint.c
+ * 
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
+ *
+ * Copyright 1999 Red Hat Software --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Checkpoint routines for the generic filesystem journaling code.  
+ * Part of the ext2fs journaling system.  
+ *
+ * Checkpointing is the process of ensuring that a section of the log is
+ * committed fully to disk, so that that portion of the log can be
+ * reused.
+ */
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+
+/*
+ * Unlink a buffer from a transaction. 
+ *
+ * Called with j_list_lock held.
+ */
+
+static inline void __buffer_unlink(struct journal_head *jh)
+{
+	transaction_t *transaction;
+
+	transaction = jh->b_cp_transaction;
+	jh->b_cp_transaction = NULL;
+
+	jh->b_cpnext->b_cpprev = jh->b_cpprev;
+	jh->b_cpprev->b_cpnext = jh->b_cpnext;
+	if (transaction->t_checkpoint_list == jh)
+		transaction->t_checkpoint_list = jh->b_cpnext;
+	if (transaction->t_checkpoint_list == jh)
+		transaction->t_checkpoint_list = NULL;
+}
+
+/*
+ * Try to release a checkpointed buffer from its transaction.
+ * Returns 1 if we released it.
+ * Requires j_list_lock
+ * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
+ */
+static int __try_to_free_cp_buf(struct journal_head *jh)
+{
+	int ret = 0;
+	struct buffer_head *bh = jh2bh(jh);
+
+	if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) {
+		JBUFFER_TRACE(jh, "remove from checkpoint list");
+		__journal_remove_checkpoint(jh);
+		jbd_unlock_bh_state(bh);
+		journal_remove_journal_head(bh);
+		BUFFER_TRACE(bh, "release");
+		__brelse(bh);
+		ret = 1;
+	} else {
+		jbd_unlock_bh_state(bh);
+	}
+	return ret;
+}
+
+/*
+ * __log_wait_for_space: wait until there is space in the journal.
+ *
+ * Called under j-state_lock *only*.  It will be unlocked if we have to wait
+ * for a checkpoint to free up some space in the log.
+ */
+void __log_wait_for_space(journal_t *journal)
+{
+	int nblocks;
+	assert_spin_locked(&journal->j_state_lock);
+
+	nblocks = jbd_space_needed(journal);
+	while (__log_space_left(journal) < nblocks) {
+		if (journal->j_flags & JFS_ABORT)
+			return;
+		spin_unlock(&journal->j_state_lock);
+		down(&journal->j_checkpoint_sem);
+
+		/*
+		 * Test again, another process may have checkpointed while we
+		 * were waiting for the checkpoint lock
+		 */
+		spin_lock(&journal->j_state_lock);
+		nblocks = jbd_space_needed(journal);
+		if (__log_space_left(journal) < nblocks) {
+			spin_unlock(&journal->j_state_lock);
+			log_do_checkpoint(journal);
+			spin_lock(&journal->j_state_lock);
+		}
+		up(&journal->j_checkpoint_sem);
+	}
+}
+
+/*
+ * We were unable to perform jbd_trylock_bh_state() inside j_list_lock.
+ * The caller must restart a list walk.  Wait for someone else to run
+ * jbd_unlock_bh_state().
+ */
+static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh)
+{
+	get_bh(bh);
+	spin_unlock(&journal->j_list_lock);
+	jbd_lock_bh_state(bh);
+	jbd_unlock_bh_state(bh);
+	put_bh(bh);
+}
+
+/*
+ * Clean up a transaction's checkpoint list.  
+ *
+ * We wait for any pending IO to complete and make sure any clean
+ * buffers are removed from the transaction. 
+ *
+ * Return 1 if we performed any actions which might have destroyed the
+ * checkpoint.  (journal_remove_checkpoint() deletes the transaction when
+ * the last checkpoint buffer is cleansed)
+ *
+ * Called with j_list_lock held.
+ */
+static int __cleanup_transaction(journal_t *journal, transaction_t *transaction)
+{
+	struct journal_head *jh, *next_jh, *last_jh;
+	struct buffer_head *bh;
+	int ret = 0;
+
+	assert_spin_locked(&journal->j_list_lock);
+	jh = transaction->t_checkpoint_list;
+	if (!jh)
+		return 0;
+
+	last_jh = jh->b_cpprev;
+	next_jh = jh;
+	do {
+		jh = next_jh;
+		bh = jh2bh(jh);
+		if (buffer_locked(bh)) {
+			atomic_inc(&bh->b_count);
+			spin_unlock(&journal->j_list_lock);
+			wait_on_buffer(bh);
+			/* the journal_head may have gone by now */
+			BUFFER_TRACE(bh, "brelse");
+			__brelse(bh);
+			goto out_return_1;
+		}
+
+		/*
+		 * This is foul
+		 */
+		if (!jbd_trylock_bh_state(bh)) {
+			jbd_sync_bh(journal, bh);
+			goto out_return_1;
+		}
+
+		if (jh->b_transaction != NULL) {
+			transaction_t *t = jh->b_transaction;
+			tid_t tid = t->t_tid;
+
+			spin_unlock(&journal->j_list_lock);
+			jbd_unlock_bh_state(bh);
+			log_start_commit(journal, tid);
+			log_wait_commit(journal, tid);
+			goto out_return_1;
+		}
+
+		/*
+		 * AKPM: I think the buffer_jbddirty test is redundant - it
+		 * shouldn't have NULL b_transaction?
+		 */
+		next_jh = jh->b_cpnext;
+		if (!buffer_dirty(bh) && !buffer_jbddirty(bh)) {
+			BUFFER_TRACE(bh, "remove from checkpoint");
+			__journal_remove_checkpoint(jh);
+			jbd_unlock_bh_state(bh);
+			journal_remove_journal_head(bh);
+			__brelse(bh);
+			ret = 1;
+		} else {
+			jbd_unlock_bh_state(bh);
+		}
+		jh = next_jh;
+	} while (jh != last_jh);
+
+	return ret;
+out_return_1:
+	spin_lock(&journal->j_list_lock);
+	return 1;
+}
+
+#define NR_BATCH	64
+
+static void
+__flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
+{
+	int i;
+
+	spin_unlock(&journal->j_list_lock);
+	ll_rw_block(WRITE, *batch_count, bhs);
+	spin_lock(&journal->j_list_lock);
+	for (i = 0; i < *batch_count; i++) {
+		struct buffer_head *bh = bhs[i];
+		clear_buffer_jwrite(bh);
+		BUFFER_TRACE(bh, "brelse");
+		__brelse(bh);
+	}
+	*batch_count = 0;
+}
+
+/*
+ * Try to flush one buffer from the checkpoint list to disk.
+ *
+ * Return 1 if something happened which requires us to abort the current
+ * scan of the checkpoint list.  
+ *
+ * Called with j_list_lock held.
+ * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
+ */
+static int __flush_buffer(journal_t *journal, struct journal_head *jh,
+			struct buffer_head **bhs, int *batch_count,
+			int *drop_count)
+{
+	struct buffer_head *bh = jh2bh(jh);
+	int ret = 0;
+
+	if (buffer_dirty(bh) && !buffer_locked(bh) && jh->b_jlist == BJ_None) {
+		J_ASSERT_JH(jh, jh->b_transaction == NULL);
+
+		/*
+		 * Important: we are about to write the buffer, and
+		 * possibly block, while still holding the journal lock.
+		 * We cannot afford to let the transaction logic start
+		 * messing around with this buffer before we write it to
+		 * disk, as that would break recoverability.  
+		 */
+		BUFFER_TRACE(bh, "queue");
+		get_bh(bh);
+		J_ASSERT_BH(bh, !buffer_jwrite(bh));
+		set_buffer_jwrite(bh);
+		bhs[*batch_count] = bh;
+		jbd_unlock_bh_state(bh);
+		(*batch_count)++;
+		if (*batch_count == NR_BATCH) {
+			__flush_batch(journal, bhs, batch_count);
+			ret = 1;
+		}
+	} else {
+		int last_buffer = 0;
+		if (jh->b_cpnext == jh) {
+			/* We may be about to drop the transaction.  Tell the
+			 * caller that the lists have changed.
+			 */
+			last_buffer = 1;
+		}
+		if (__try_to_free_cp_buf(jh)) {
+			(*drop_count)++;
+			ret = last_buffer;
+		}
+	}
+	return ret;
+}
+
+/*
+ * Perform an actual checkpoint.  We don't write out only enough to
+ * satisfy the current blocked requests: rather we submit a reasonably
+ * sized chunk of the outstanding data to disk at once for
+ * efficiency.  __log_wait_for_space() will retry if we didn't free enough.
+ * 
+ * However, we _do_ take into account the amount requested so that once
+ * the IO has been queued, we can return as soon as enough of it has
+ * completed to disk.  
+ *
+ * The journal should be locked before calling this function.
+ */
+int log_do_checkpoint(journal_t *journal)
+{
+	int result;
+	int batch_count = 0;
+	struct buffer_head *bhs[NR_BATCH];
+
+	jbd_debug(1, "Start checkpoint\n");
+
+	/* 
+	 * First thing: if there are any transactions in the log which
+	 * don't need checkpointing, just eliminate them from the
+	 * journal straight away.  
+	 */
+	result = cleanup_journal_tail(journal);
+	jbd_debug(1, "cleanup_journal_tail returned %d\n", result);
+	if (result <= 0)
+		return result;
+
+	/*
+	 * OK, we need to start writing disk blocks.  Try to free up a
+	 * quarter of the log in a single checkpoint if we can.
+	 */
+	/*
+	 * AKPM: check this code.  I had a feeling a while back that it
+	 * degenerates into a busy loop at unmount time.
+	 */
+	spin_lock(&journal->j_list_lock);
+	while (journal->j_checkpoint_transactions) {
+		transaction_t *transaction;
+		struct journal_head *jh, *last_jh, *next_jh;
+		int drop_count = 0;
+		int cleanup_ret, retry = 0;
+		tid_t this_tid;
+
+		transaction = journal->j_checkpoint_transactions;
+		this_tid = transaction->t_tid;
+		jh = transaction->t_checkpoint_list;
+		last_jh = jh->b_cpprev;
+		next_jh = jh;
+		do {
+			struct buffer_head *bh;
+
+			jh = next_jh;
+			next_jh = jh->b_cpnext;
+			bh = jh2bh(jh);
+			if (!jbd_trylock_bh_state(bh)) {
+				jbd_sync_bh(journal, bh);
+				spin_lock(&journal->j_list_lock);
+				retry = 1;
+				break;
+			}
+			retry = __flush_buffer(journal, jh, bhs, &batch_count, &drop_count);
+			if (cond_resched_lock(&journal->j_list_lock)) {
+				retry = 1;
+				break;
+			}
+		} while (jh != last_jh && !retry);
+
+		if (batch_count)
+			__flush_batch(journal, bhs, &batch_count);
+
+		/*
+		 * If someone cleaned up this transaction while we slept, we're
+		 * done
+		 */
+		if (journal->j_checkpoint_transactions != transaction)
+			break;
+		if (retry)
+			continue;
+		/*
+		 * Maybe it's a new transaction, but it fell at the same
+		 * address
+		 */
+		if (transaction->t_tid != this_tid)
+			continue;
+		/*
+		 * We have walked the whole transaction list without
+		 * finding anything to write to disk.  We had better be
+		 * able to make some progress or we are in trouble. 
+		 */
+		cleanup_ret = __cleanup_transaction(journal, transaction);
+		J_ASSERT(drop_count != 0 || cleanup_ret != 0);
+		if (journal->j_checkpoint_transactions != transaction)
+			break;
+	}
+	spin_unlock(&journal->j_list_lock);
+	result = cleanup_journal_tail(journal);
+	if (result < 0)
+		return result;
+
+	return 0;
+}
+
+/*
+ * Check the list of checkpoint transactions for the journal to see if
+ * we have already got rid of any since the last update of the log tail
+ * in the journal superblock.  If so, we can instantly roll the
+ * superblock forward to remove those transactions from the log.
+ * 
+ * Return <0 on error, 0 on success, 1 if there was nothing to clean up.
+ * 
+ * Called with the journal lock held.
+ *
+ * This is the only part of the journaling code which really needs to be
+ * aware of transaction aborts.  Checkpointing involves writing to the
+ * main filesystem area rather than to the journal, so it can proceed
+ * even in abort state, but we must not update the journal superblock if
+ * we have an abort error outstanding.
+ */
+
+int cleanup_journal_tail(journal_t *journal)
+{
+	transaction_t * transaction;
+	tid_t		first_tid;
+	unsigned long	blocknr, freed;
+
+	/* OK, work out the oldest transaction remaining in the log, and
+	 * the log block it starts at. 
+	 * 
+	 * If the log is now empty, we need to work out which is the
+	 * next transaction ID we will write, and where it will
+	 * start. */
+
+	spin_lock(&journal->j_state_lock);
+	spin_lock(&journal->j_list_lock);
+	transaction = journal->j_checkpoint_transactions;
+	if (transaction) {
+		first_tid = transaction->t_tid;
+		blocknr = transaction->t_log_start;
+	} else if ((transaction = journal->j_committing_transaction) != NULL) {
+		first_tid = transaction->t_tid;
+		blocknr = transaction->t_log_start;
+	} else if ((transaction = journal->j_running_transaction) != NULL) {
+		first_tid = transaction->t_tid;
+		blocknr = journal->j_head;
+	} else {
+		first_tid = journal->j_transaction_sequence;
+		blocknr = journal->j_head;
+	}
+	spin_unlock(&journal->j_list_lock);
+	J_ASSERT(blocknr != 0);
+
+	/* If the oldest pinned transaction is at the tail of the log
+           already then there's not much we can do right now. */
+	if (journal->j_tail_sequence == first_tid) {
+		spin_unlock(&journal->j_state_lock);
+		return 1;
+	}
+
+	/* OK, update the superblock to recover the freed space.
+	 * Physical blocks come first: have we wrapped beyond the end of
+	 * the log?  */
+	freed = blocknr - journal->j_tail;
+	if (blocknr < journal->j_tail)
+		freed = freed + journal->j_last - journal->j_first;
+
+	jbd_debug(1,
+		  "Cleaning journal tail from %d to %d (offset %lu), "
+		  "freeing %lu\n",
+		  journal->j_tail_sequence, first_tid, blocknr, freed);
+
+	journal->j_free += freed;
+	journal->j_tail_sequence = first_tid;
+	journal->j_tail = blocknr;
+	spin_unlock(&journal->j_state_lock);
+	if (!(journal->j_flags & JFS_ABORT))
+		journal_update_superblock(journal, 1);
+	return 0;
+}
+
+
+/* Checkpoint list management */
+
+/*
+ * journal_clean_checkpoint_list
+ *
+ * Find all the written-back checkpoint buffers in the journal and release them.
+ *
+ * Called with the journal locked.
+ * Called with j_list_lock held.
+ * Returns number of bufers reaped (for debug)
+ */
+
+int __journal_clean_checkpoint_list(journal_t *journal)
+{
+	transaction_t *transaction, *last_transaction, *next_transaction;
+	int ret = 0;
+
+	transaction = journal->j_checkpoint_transactions;
+	if (transaction == 0)
+		goto out;
+
+	last_transaction = transaction->t_cpprev;
+	next_transaction = transaction;
+	do {
+		struct journal_head *jh;
+
+		transaction = next_transaction;
+		next_transaction = transaction->t_cpnext;
+		jh = transaction->t_checkpoint_list;
+		if (jh) {
+			struct journal_head *last_jh = jh->b_cpprev;
+			struct journal_head *next_jh = jh;
+
+			do {
+				jh = next_jh;
+				next_jh = jh->b_cpnext;
+				/* Use trylock because of the ranknig */
+				if (jbd_trylock_bh_state(jh2bh(jh)))
+					ret += __try_to_free_cp_buf(jh);
+				/*
+				 * This function only frees up some memory
+				 * if possible so we dont have an obligation
+				 * to finish processing. Bail out if preemption
+				 * requested:
+				 */
+				if (need_resched())
+					goto out;
+			} while (jh != last_jh);
+		}
+	} while (transaction != last_transaction);
+out:
+	return ret;
+}
+
+/* 
+ * journal_remove_checkpoint: called after a buffer has been committed
+ * to disk (either by being write-back flushed to disk, or being
+ * committed to the log).
+ *
+ * We cannot safely clean a transaction out of the log until all of the
+ * buffer updates committed in that transaction have safely been stored
+ * elsewhere on disk.  To achieve this, all of the buffers in a
+ * transaction need to be maintained on the transaction's checkpoint
+ * list until they have been rewritten, at which point this function is
+ * called to remove the buffer from the existing transaction's
+ * checkpoint list.  
+ *
+ * This function is called with the journal locked.
+ * This function is called with j_list_lock held.
+ */
+
+void __journal_remove_checkpoint(struct journal_head *jh)
+{
+	transaction_t *transaction;
+	journal_t *journal;
+
+	JBUFFER_TRACE(jh, "entry");
+
+	if ((transaction = jh->b_cp_transaction) == NULL) {
+		JBUFFER_TRACE(jh, "not on transaction");
+		goto out;
+	}
+	journal = transaction->t_journal;
+
+	__buffer_unlink(jh);
+
+	if (transaction->t_checkpoint_list != NULL)
+		goto out;
+	JBUFFER_TRACE(jh, "transaction has no more buffers");
+
+	/*
+	 * There is one special case to worry about: if we have just pulled the
+	 * buffer off a committing transaction's forget list, then even if the
+	 * checkpoint list is empty, the transaction obviously cannot be
+	 * dropped!
+	 *
+	 * The locking here around j_committing_transaction is a bit sleazy.
+	 * See the comment at the end of journal_commit_transaction().
+	 */
+	if (transaction == journal->j_committing_transaction) {
+		JBUFFER_TRACE(jh, "belongs to committing transaction");
+		goto out;
+	}
+
+	/* OK, that was the last buffer for the transaction: we can now
+	   safely remove this transaction from the log */
+
+	__journal_drop_transaction(journal, transaction);
+
+	/* Just in case anybody was waiting for more transactions to be
+           checkpointed... */
+	wake_up(&journal->j_wait_logspace);
+out:
+	JBUFFER_TRACE(jh, "exit");
+}
+
+/*
+ * journal_insert_checkpoint: put a committed buffer onto a checkpoint
+ * list so that we know when it is safe to clean the transaction out of
+ * the log.
+ *
+ * Called with the journal locked.
+ * Called with j_list_lock held.
+ */
+void __journal_insert_checkpoint(struct journal_head *jh, 
+			       transaction_t *transaction)
+{
+	JBUFFER_TRACE(jh, "entry");
+	J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh)));
+	J_ASSERT_JH(jh, jh->b_cp_transaction == NULL);
+
+	jh->b_cp_transaction = transaction;
+
+	if (!transaction->t_checkpoint_list) {
+		jh->b_cpnext = jh->b_cpprev = jh;
+	} else {
+		jh->b_cpnext = transaction->t_checkpoint_list;
+		jh->b_cpprev = transaction->t_checkpoint_list->b_cpprev;
+		jh->b_cpprev->b_cpnext = jh;
+		jh->b_cpnext->b_cpprev = jh;
+	}
+	transaction->t_checkpoint_list = jh;
+}
+
+/*
+ * We've finished with this transaction structure: adios...
+ * 
+ * The transaction must have no links except for the checkpoint by this
+ * point.
+ *
+ * Called with the journal locked.
+ * Called with j_list_lock held.
+ */
+
+void __journal_drop_transaction(journal_t *journal, transaction_t *transaction)
+{
+	assert_spin_locked(&journal->j_list_lock);
+	if (transaction->t_cpnext) {
+		transaction->t_cpnext->t_cpprev = transaction->t_cpprev;
+		transaction->t_cpprev->t_cpnext = transaction->t_cpnext;
+		if (journal->j_checkpoint_transactions == transaction)
+			journal->j_checkpoint_transactions =
+				transaction->t_cpnext;
+		if (journal->j_checkpoint_transactions == transaction)
+			journal->j_checkpoint_transactions = NULL;
+	}
+
+	J_ASSERT(transaction->t_state == T_FINISHED);
+	J_ASSERT(transaction->t_buffers == NULL);
+	J_ASSERT(transaction->t_sync_datalist == NULL);
+	J_ASSERT(transaction->t_forget == NULL);
+	J_ASSERT(transaction->t_iobuf_list == NULL);
+	J_ASSERT(transaction->t_shadow_list == NULL);
+	J_ASSERT(transaction->t_log_list == NULL);
+	J_ASSERT(transaction->t_checkpoint_list == NULL);
+	J_ASSERT(transaction->t_updates == 0);
+	J_ASSERT(journal->j_committing_transaction != transaction);
+	J_ASSERT(journal->j_running_transaction != transaction);
+
+	jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid);
+	kfree(transaction);
+}
diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
new file mode 100644
index 00000000000..dac720c837a
--- /dev/null
+++ b/fs/jbd/commit.c
@@ -0,0 +1,844 @@
+/*
+ * linux/fs/commit.c
+ *
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
+ *
+ * Copyright 1998 Red Hat corp --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Journal commit routines for the generic filesystem journaling code;
+ * part of the ext2fs journaling system.
+ */
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/pagemap.h>
+#include <linux/smp_lock.h>
+
+/*
+ * Default IO end handler for temporary BJ_IO buffer_heads.
+ */
+static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate)
+{
+	BUFFER_TRACE(bh, "");
+	if (uptodate)
+		set_buffer_uptodate(bh);
+	else
+		clear_buffer_uptodate(bh);
+	unlock_buffer(bh);
+}
+
+/*
+ * When an ext3-ordered file is truncated, it is possible that many pages are
+ * not sucessfully freed, because they are attached to a committing transaction.
+ * After the transaction commits, these pages are left on the LRU, with no
+ * ->mapping, and with attached buffers.  These pages are trivially reclaimable
+ * by the VM, but their apparent absence upsets the VM accounting, and it makes
+ * the numbers in /proc/meminfo look odd.
+ *
+ * So here, we have a buffer which has just come off the forget list.  Look to
+ * see if we can strip all buffers from the backing page.
+ *
+ * Called under lock_journal(), and possibly under journal_datalist_lock.  The
+ * caller provided us with a ref against the buffer, and we drop that here.
+ */
+static void release_buffer_page(struct buffer_head *bh)
+{
+	struct page *page;
+
+	if (buffer_dirty(bh))
+		goto nope;
+	if (atomic_read(&bh->b_count) != 1)
+		goto nope;
+	page = bh->b_page;
+	if (!page)
+		goto nope;
+	if (page->mapping)
+		goto nope;
+
+	/* OK, it's a truncated page */
+	if (TestSetPageLocked(page))
+		goto nope;
+
+	page_cache_get(page);
+	__brelse(bh);
+	try_to_free_buffers(page);
+	unlock_page(page);
+	page_cache_release(page);
+	return;
+
+nope:
+	__brelse(bh);
+}
+
+/*
+ * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is
+ * held.  For ranking reasons we must trylock.  If we lose, schedule away and
+ * return 0.  j_list_lock is dropped in this case.
+ */
+static int inverted_lock(journal_t *journal, struct buffer_head *bh)
+{
+	if (!jbd_trylock_bh_state(bh)) {
+		spin_unlock(&journal->j_list_lock);
+		schedule();
+		return 0;
+	}
+	return 1;
+}
+
+/* Done it all: now write the commit record.  We should have
+ * cleaned up our previous buffers by now, so if we are in abort
+ * mode we can now just skip the rest of the journal write
+ * entirely.
+ *
+ * Returns 1 if the journal needs to be aborted or 0 on success
+ */
+static int journal_write_commit_record(journal_t *journal,
+					transaction_t *commit_transaction)
+{
+	struct journal_head *descriptor;
+	struct buffer_head *bh;
+	int i, ret;
+	int barrier_done = 0;
+
+	if (is_journal_aborted(journal))
+		return 0;
+
+	descriptor = journal_get_descriptor_buffer(journal);
+	if (!descriptor)
+		return 1;
+
+	bh = jh2bh(descriptor);
+
+	/* AKPM: buglet - add `i' to tmp! */
+	for (i = 0; i < bh->b_size; i += 512) {
+		journal_header_t *tmp = (journal_header_t*)bh->b_data;
+		tmp->h_magic = cpu_to_be32(JFS_MAGIC_NUMBER);
+		tmp->h_blocktype = cpu_to_be32(JFS_COMMIT_BLOCK);
+		tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid);
+	}
+
+	JBUFFER_TRACE(descriptor, "write commit block");
+	set_buffer_dirty(bh);
+	if (journal->j_flags & JFS_BARRIER) {
+		set_buffer_ordered(bh);
+		barrier_done = 1;
+	}
+	ret = sync_dirty_buffer(bh);
+	/* is it possible for another commit to fail at roughly
+	 * the same time as this one?  If so, we don't want to
+	 * trust the barrier flag in the super, but instead want
+	 * to remember if we sent a barrier request
+	 */
+	if (ret == -EOPNOTSUPP && barrier_done) {
+		char b[BDEVNAME_SIZE];
+
+		printk(KERN_WARNING
+			"JBD: barrier-based sync failed on %s - "
+			"disabling barriers\n",
+			bdevname(journal->j_dev, b));
+		spin_lock(&journal->j_state_lock);
+		journal->j_flags &= ~JFS_BARRIER;
+		spin_unlock(&journal->j_state_lock);
+
+		/* And try again, without the barrier */
+		clear_buffer_ordered(bh);
+		set_buffer_uptodate(bh);
+		set_buffer_dirty(bh);
+		ret = sync_dirty_buffer(bh);
+	}
+	put_bh(bh);		/* One for getblk() */
+	journal_put_journal_head(descriptor);
+
+	return (ret == -EIO);
+}
+
+/*
+ * journal_commit_transaction
+ *
+ * The primary function for committing a transaction to the log.  This
+ * function is called by the journal thread to begin a complete commit.
+ */
+void journal_commit_transaction(journal_t *journal)
+{
+	transaction_t *commit_transaction;
+	struct journal_head *jh, *new_jh, *descriptor;
+	struct buffer_head **wbuf = journal->j_wbuf;
+	int bufs;
+	int flags;
+	int err;
+	unsigned long blocknr;
+	char *tagp = NULL;
+	journal_header_t *header;
+	journal_block_tag_t *tag = NULL;
+	int space_left = 0;
+	int first_tag = 0;
+	int tag_flag;
+	int i;
+
+	/*
+	 * First job: lock down the current transaction and wait for
+	 * all outstanding updates to complete.
+	 */
+
+#ifdef COMMIT_STATS
+	spin_lock(&journal->j_list_lock);
+	summarise_journal_usage(journal);
+	spin_unlock(&journal->j_list_lock);
+#endif
+
+	/* Do we need to erase the effects of a prior journal_flush? */
+	if (journal->j_flags & JFS_FLUSHED) {
+		jbd_debug(3, "super block updated\n");
+		journal_update_superblock(journal, 1);
+	} else {
+		jbd_debug(3, "superblock not updated\n");
+	}
+
+	J_ASSERT(journal->j_running_transaction != NULL);
+	J_ASSERT(journal->j_committing_transaction == NULL);
+
+	commit_transaction = journal->j_running_transaction;
+	J_ASSERT(commit_transaction->t_state == T_RUNNING);
+
+	jbd_debug(1, "JBD: starting commit of transaction %d\n",
+			commit_transaction->t_tid);
+
+	spin_lock(&journal->j_state_lock);
+	commit_transaction->t_state = T_LOCKED;
+
+	spin_lock(&commit_transaction->t_handle_lock);
+	while (commit_transaction->t_updates) {
+		DEFINE_WAIT(wait);
+
+		prepare_to_wait(&journal->j_wait_updates, &wait,
+					TASK_UNINTERRUPTIBLE);
+		if (commit_transaction->t_updates) {
+			spin_unlock(&commit_transaction->t_handle_lock);
+			spin_unlock(&journal->j_state_lock);
+			schedule();
+			spin_lock(&journal->j_state_lock);
+			spin_lock(&commit_transaction->t_handle_lock);
+		}
+		finish_wait(&journal->j_wait_updates, &wait);
+	}
+	spin_unlock(&commit_transaction->t_handle_lock);
+
+	J_ASSERT (commit_transaction->t_outstanding_credits <=
+			journal->j_max_transaction_buffers);
+
+	/*
+	 * First thing we are allowed to do is to discard any remaining
+	 * BJ_Reserved buffers.  Note, it is _not_ permissible to assume
+	 * that there are no such buffers: if a large filesystem
+	 * operation like a truncate needs to split itself over multiple
+	 * transactions, then it may try to do a journal_restart() while
+	 * there are still BJ_Reserved buffers outstanding.  These must
+	 * be released cleanly from the current transaction.
+	 *
+	 * In this case, the filesystem must still reserve write access
+	 * again before modifying the buffer in the new transaction, but
+	 * we do not require it to remember exactly which old buffers it
+	 * has reserved.  This is consistent with the existing behaviour
+	 * that multiple journal_get_write_access() calls to the same
+	 * buffer are perfectly permissable.
+	 */
+	while (commit_transaction->t_reserved_list) {
+		jh = commit_transaction->t_reserved_list;
+		JBUFFER_TRACE(jh, "reserved, unused: refile");
+		/*
+		 * A journal_get_undo_access()+journal_release_buffer() may
+		 * leave undo-committed data.
+		 */
+		if (jh->b_committed_data) {
+			struct buffer_head *bh = jh2bh(jh);
+
+			jbd_lock_bh_state(bh);
+			if (jh->b_committed_data) {
+				kfree(jh->b_committed_data);
+				jh->b_committed_data = NULL;
+			}
+			jbd_unlock_bh_state(bh);
+		}
+		journal_refile_buffer(journal, jh);
+	}
+
+	/*
+	 * Now try to drop any written-back buffers from the journal's
+	 * checkpoint lists.  We do this *before* commit because it potentially
+	 * frees some memory
+	 */
+	spin_lock(&journal->j_list_lock);
+	__journal_clean_checkpoint_list(journal);
+	spin_unlock(&journal->j_list_lock);
+
+	jbd_debug (3, "JBD: commit phase 1\n");
+
+	/*
+	 * Switch to a new revoke table.
+	 */
+	journal_switch_revoke_table(journal);
+
+	commit_transaction->t_state = T_FLUSH;
+	journal->j_committing_transaction = commit_transaction;
+	journal->j_running_transaction = NULL;
+	commit_transaction->t_log_start = journal->j_head;
+	wake_up(&journal->j_wait_transaction_locked);
+	spin_unlock(&journal->j_state_lock);
+
+	jbd_debug (3, "JBD: commit phase 2\n");
+
+	/*
+	 * First, drop modified flag: all accesses to the buffers
+	 * will be tracked for a new trasaction only -bzzz
+	 */
+	spin_lock(&journal->j_list_lock);
+	if (commit_transaction->t_buffers) {
+		new_jh = jh = commit_transaction->t_buffers->b_tnext;
+		do {
+			J_ASSERT_JH(new_jh, new_jh->b_modified == 1 ||
+					new_jh->b_modified == 0);
+			new_jh->b_modified = 0;
+			new_jh = new_jh->b_tnext;
+		} while (new_jh != jh);
+	}
+	spin_unlock(&journal->j_list_lock);
+
+	/*
+	 * Now start flushing things to disk, in the order they appear
+	 * on the transaction lists.  Data blocks go first.
+	 */
+
+	err = 0;
+	/*
+	 * Whenever we unlock the journal and sleep, things can get added
+	 * onto ->t_sync_datalist, so we have to keep looping back to
+	 * write_out_data until we *know* that the list is empty.
+	 */
+	bufs = 0;
+	/*
+	 * Cleanup any flushed data buffers from the data list.  Even in
+	 * abort mode, we want to flush this out as soon as possible.
+	 */
+write_out_data:
+	cond_resched();
+	spin_lock(&journal->j_list_lock);
+
+	while (commit_transaction->t_sync_datalist) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_sync_datalist;
+		commit_transaction->t_sync_datalist = jh->b_tnext;
+		bh = jh2bh(jh);
+		if (buffer_locked(bh)) {
+			BUFFER_TRACE(bh, "locked");
+			if (!inverted_lock(journal, bh))
+				goto write_out_data;
+			__journal_temp_unlink_buffer(jh);
+			__journal_file_buffer(jh, commit_transaction,
+						BJ_Locked);
+			jbd_unlock_bh_state(bh);
+			if (lock_need_resched(&journal->j_list_lock)) {
+				spin_unlock(&journal->j_list_lock);
+				goto write_out_data;
+			}
+		} else {
+			if (buffer_dirty(bh)) {
+				BUFFER_TRACE(bh, "start journal writeout");
+				get_bh(bh);
+				wbuf[bufs++] = bh;
+				if (bufs == journal->j_wbufsize) {
+					jbd_debug(2, "submit %d writes\n",
+							bufs);
+					spin_unlock(&journal->j_list_lock);
+					ll_rw_block(WRITE, bufs, wbuf);
+					journal_brelse_array(wbuf, bufs);
+					bufs = 0;
+					goto write_out_data;
+				}
+			} else {
+				BUFFER_TRACE(bh, "writeout complete: unfile");
+				if (!inverted_lock(journal, bh))
+					goto write_out_data;
+				__journal_unfile_buffer(jh);
+				jbd_unlock_bh_state(bh);
+				journal_remove_journal_head(bh);
+				put_bh(bh);
+				if (lock_need_resched(&journal->j_list_lock)) {
+					spin_unlock(&journal->j_list_lock);
+					goto write_out_data;
+				}
+			}
+		}
+	}
+
+	if (bufs) {
+		spin_unlock(&journal->j_list_lock);
+		ll_rw_block(WRITE, bufs, wbuf);
+		journal_brelse_array(wbuf, bufs);
+		spin_lock(&journal->j_list_lock);
+	}
+
+	/*
+	 * Wait for all previously submitted IO to complete.
+	 */
+	while (commit_transaction->t_locked_list) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_locked_list->b_tprev;
+		bh = jh2bh(jh);
+		get_bh(bh);
+		if (buffer_locked(bh)) {
+			spin_unlock(&journal->j_list_lock);
+			wait_on_buffer(bh);
+			if (unlikely(!buffer_uptodate(bh)))
+				err = -EIO;
+			spin_lock(&journal->j_list_lock);
+		}
+		if (!inverted_lock(journal, bh)) {
+			put_bh(bh);
+			spin_lock(&journal->j_list_lock);
+			continue;
+		}
+		if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) {
+			__journal_unfile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+			journal_remove_journal_head(bh);
+			put_bh(bh);
+		} else {
+			jbd_unlock_bh_state(bh);
+		}
+		put_bh(bh);
+		cond_resched_lock(&journal->j_list_lock);
+	}
+	spin_unlock(&journal->j_list_lock);
+
+	if (err)
+		__journal_abort_hard(journal);
+
+	journal_write_revoke_records(journal, commit_transaction);
+
+	jbd_debug(3, "JBD: commit phase 2\n");
+
+	/*
+	 * If we found any dirty or locked buffers, then we should have
+	 * looped back up to the write_out_data label.  If there weren't
+	 * any then journal_clean_data_list should have wiped the list
+	 * clean by now, so check that it is in fact empty.
+	 */
+	J_ASSERT (commit_transaction->t_sync_datalist == NULL);
+
+	jbd_debug (3, "JBD: commit phase 3\n");
+
+	/*
+	 * Way to go: we have now written out all of the data for a
+	 * transaction!  Now comes the tricky part: we need to write out
+	 * metadata.  Loop over the transaction's entire buffer list:
+	 */
+	commit_transaction->t_state = T_COMMIT;
+
+	descriptor = NULL;
+	bufs = 0;
+	while (commit_transaction->t_buffers) {
+
+		/* Find the next buffer to be journaled... */
+
+		jh = commit_transaction->t_buffers;
+
+		/* If we're in abort mode, we just un-journal the buffer and
+		   release it for background writing. */
+
+		if (is_journal_aborted(journal)) {
+			JBUFFER_TRACE(jh, "journal is aborting: refile");
+			journal_refile_buffer(journal, jh);
+			/* If that was the last one, we need to clean up
+			 * any descriptor buffers which may have been
+			 * already allocated, even if we are now
+			 * aborting. */
+			if (!commit_transaction->t_buffers)
+				goto start_journal_io;
+			continue;
+		}
+
+		/* Make sure we have a descriptor block in which to
+		   record the metadata buffer. */
+
+		if (!descriptor) {
+			struct buffer_head *bh;
+
+			J_ASSERT (bufs == 0);
+
+			jbd_debug(4, "JBD: get descriptor\n");
+
+			descriptor = journal_get_descriptor_buffer(journal);
+			if (!descriptor) {
+				__journal_abort_hard(journal);
+				continue;
+			}
+
+			bh = jh2bh(descriptor);
+			jbd_debug(4, "JBD: got buffer %llu (%p)\n",
+				(unsigned long long)bh->b_blocknr, bh->b_data);
+			header = (journal_header_t *)&bh->b_data[0];
+			header->h_magic     = cpu_to_be32(JFS_MAGIC_NUMBER);
+			header->h_blocktype = cpu_to_be32(JFS_DESCRIPTOR_BLOCK);
+			header->h_sequence  = cpu_to_be32(commit_transaction->t_tid);
+
+			tagp = &bh->b_data[sizeof(journal_header_t)];
+			space_left = bh->b_size - sizeof(journal_header_t);
+			first_tag = 1;
+			set_buffer_jwrite(bh);
+			set_buffer_dirty(bh);
+			wbuf[bufs++] = bh;
+
+			/* Record it so that we can wait for IO
+                           completion later */
+			BUFFER_TRACE(bh, "ph3: file as descriptor");
+			journal_file_buffer(descriptor, commit_transaction,
+					BJ_LogCtl);
+		}
+
+		/* Where is the buffer to be written? */
+
+		err = journal_next_log_block(journal, &blocknr);
+		/* If the block mapping failed, just abandon the buffer
+		   and repeat this loop: we'll fall into the
+		   refile-on-abort condition above. */
+		if (err) {
+			__journal_abort_hard(journal);
+			continue;
+		}
+
+		/*
+		 * start_this_handle() uses t_outstanding_credits to determine
+		 * the free space in the log, but this counter is changed
+		 * by journal_next_log_block() also.
+		 */
+		commit_transaction->t_outstanding_credits--;
+
+		/* Bump b_count to prevent truncate from stumbling over
+                   the shadowed buffer!  @@@ This can go if we ever get
+                   rid of the BJ_IO/BJ_Shadow pairing of buffers. */
+		atomic_inc(&jh2bh(jh)->b_count);
+
+		/* Make a temporary IO buffer with which to write it out
+                   (this will requeue both the metadata buffer and the
+                   temporary IO buffer). new_bh goes on BJ_IO*/
+
+		set_bit(BH_JWrite, &jh2bh(jh)->b_state);
+		/*
+		 * akpm: journal_write_metadata_buffer() sets
+		 * new_bh->b_transaction to commit_transaction.
+		 * We need to clean this up before we release new_bh
+		 * (which is of type BJ_IO)
+		 */
+		JBUFFER_TRACE(jh, "ph3: write metadata");
+		flags = journal_write_metadata_buffer(commit_transaction,
+						      jh, &new_jh, blocknr);
+		set_bit(BH_JWrite, &jh2bh(new_jh)->b_state);
+		wbuf[bufs++] = jh2bh(new_jh);
+
+		/* Record the new block's tag in the current descriptor
+                   buffer */
+
+		tag_flag = 0;
+		if (flags & 1)
+			tag_flag |= JFS_FLAG_ESCAPE;
+		if (!first_tag)
+			tag_flag |= JFS_FLAG_SAME_UUID;
+
+		tag = (journal_block_tag_t *) tagp;
+		tag->t_blocknr = cpu_to_be32(jh2bh(jh)->b_blocknr);
+		tag->t_flags = cpu_to_be32(tag_flag);
+		tagp += sizeof(journal_block_tag_t);
+		space_left -= sizeof(journal_block_tag_t);
+
+		if (first_tag) {
+			memcpy (tagp, journal->j_uuid, 16);
+			tagp += 16;
+			space_left -= 16;
+			first_tag = 0;
+		}
+
+		/* If there's no more to do, or if the descriptor is full,
+		   let the IO rip! */
+
+		if (bufs == journal->j_wbufsize ||
+		    commit_transaction->t_buffers == NULL ||
+		    space_left < sizeof(journal_block_tag_t) + 16) {
+
+			jbd_debug(4, "JBD: Submit %d IOs\n", bufs);
+
+			/* Write an end-of-descriptor marker before
+                           submitting the IOs.  "tag" still points to
+                           the last tag we set up. */
+
+			tag->t_flags |= cpu_to_be32(JFS_FLAG_LAST_TAG);
+
+start_journal_io:
+			for (i = 0; i < bufs; i++) {
+				struct buffer_head *bh = wbuf[i];
+				lock_buffer(bh);
+				clear_buffer_dirty(bh);
+				set_buffer_uptodate(bh);
+				bh->b_end_io = journal_end_buffer_io_sync;
+				submit_bh(WRITE, bh);
+			}
+			cond_resched();
+
+			/* Force a new descriptor to be generated next
+                           time round the loop. */
+			descriptor = NULL;
+			bufs = 0;
+		}
+	}
+
+	/* Lo and behold: we have just managed to send a transaction to
+           the log.  Before we can commit it, wait for the IO so far to
+           complete.  Control buffers being written are on the
+           transaction's t_log_list queue, and metadata buffers are on
+           the t_iobuf_list queue.
+
+	   Wait for the buffers in reverse order.  That way we are
+	   less likely to be woken up until all IOs have completed, and
+	   so we incur less scheduling load.
+	*/
+
+	jbd_debug(3, "JBD: commit phase 4\n");
+
+	/*
+	 * akpm: these are BJ_IO, and j_list_lock is not needed.
+	 * See __journal_try_to_free_buffer.
+	 */
+wait_for_iobuf:
+	while (commit_transaction->t_iobuf_list != NULL) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_iobuf_list->b_tprev;
+		bh = jh2bh(jh);
+		if (buffer_locked(bh)) {
+			wait_on_buffer(bh);
+			goto wait_for_iobuf;
+		}
+		if (cond_resched())
+			goto wait_for_iobuf;
+
+		if (unlikely(!buffer_uptodate(bh)))
+			err = -EIO;
+
+		clear_buffer_jwrite(bh);
+
+		JBUFFER_TRACE(jh, "ph4: unfile after journal write");
+		journal_unfile_buffer(journal, jh);
+
+		/*
+		 * ->t_iobuf_list should contain only dummy buffer_heads
+		 * which were created by journal_write_metadata_buffer().
+		 */
+		BUFFER_TRACE(bh, "dumping temporary bh");
+		journal_put_journal_head(jh);
+		__brelse(bh);
+		J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0);
+		free_buffer_head(bh);
+
+		/* We also have to unlock and free the corresponding
+                   shadowed buffer */
+		jh = commit_transaction->t_shadow_list->b_tprev;
+		bh = jh2bh(jh);
+		clear_bit(BH_JWrite, &bh->b_state);
+		J_ASSERT_BH(bh, buffer_jbddirty(bh));
+
+		/* The metadata is now released for reuse, but we need
+                   to remember it against this transaction so that when
+                   we finally commit, we can do any checkpointing
+                   required. */
+		JBUFFER_TRACE(jh, "file as BJ_Forget");
+		journal_file_buffer(jh, commit_transaction, BJ_Forget);
+		/* Wake up any transactions which were waiting for this
+		   IO to complete */
+		wake_up_bit(&bh->b_state, BH_Unshadow);
+		JBUFFER_TRACE(jh, "brelse shadowed buffer");
+		__brelse(bh);
+	}
+
+	J_ASSERT (commit_transaction->t_shadow_list == NULL);
+
+	jbd_debug(3, "JBD: commit phase 5\n");
+
+	/* Here we wait for the revoke record and descriptor record buffers */
+ wait_for_ctlbuf:
+	while (commit_transaction->t_log_list != NULL) {
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_log_list->b_tprev;
+		bh = jh2bh(jh);
+		if (buffer_locked(bh)) {
+			wait_on_buffer(bh);
+			goto wait_for_ctlbuf;
+		}
+		if (cond_resched())
+			goto wait_for_ctlbuf;
+
+		if (unlikely(!buffer_uptodate(bh)))
+			err = -EIO;
+
+		BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile");
+		clear_buffer_jwrite(bh);
+		journal_unfile_buffer(journal, jh);
+		journal_put_journal_head(jh);
+		__brelse(bh);		/* One for getblk */
+		/* AKPM: bforget here */
+	}
+
+	jbd_debug(3, "JBD: commit phase 6\n");
+
+	if (journal_write_commit_record(journal, commit_transaction))
+		err = -EIO;
+
+	if (err)
+		__journal_abort_hard(journal);
+
+	/* End of a transaction!  Finally, we can do checkpoint
+           processing: any buffers committed as a result of this
+           transaction can be removed from any checkpoint list it was on
+           before. */
+
+	jbd_debug(3, "JBD: commit phase 7\n");
+
+	J_ASSERT(commit_transaction->t_sync_datalist == NULL);
+	J_ASSERT(commit_transaction->t_buffers == NULL);
+	J_ASSERT(commit_transaction->t_checkpoint_list == NULL);
+	J_ASSERT(commit_transaction->t_iobuf_list == NULL);
+	J_ASSERT(commit_transaction->t_shadow_list == NULL);
+	J_ASSERT(commit_transaction->t_log_list == NULL);
+
+restart_loop:
+	while (commit_transaction->t_forget) {
+		transaction_t *cp_transaction;
+		struct buffer_head *bh;
+
+		jh = commit_transaction->t_forget;
+		bh = jh2bh(jh);
+		jbd_lock_bh_state(bh);
+		J_ASSERT_JH(jh,	jh->b_transaction == commit_transaction ||
+			jh->b_transaction == journal->j_running_transaction);
+
+		/*
+		 * If there is undo-protected committed data against
+		 * this buffer, then we can remove it now.  If it is a
+		 * buffer needing such protection, the old frozen_data
+		 * field now points to a committed version of the
+		 * buffer, so rotate that field to the new committed
+		 * data.
+		 *
+		 * Otherwise, we can just throw away the frozen data now.
+		 */
+		if (jh->b_committed_data) {
+			kfree(jh->b_committed_data);
+			jh->b_committed_data = NULL;
+			if (jh->b_frozen_data) {
+				jh->b_committed_data = jh->b_frozen_data;
+				jh->b_frozen_data = NULL;
+			}
+		} else if (jh->b_frozen_data) {
+			kfree(jh->b_frozen_data);
+			jh->b_frozen_data = NULL;
+		}
+
+		spin_lock(&journal->j_list_lock);
+		cp_transaction = jh->b_cp_transaction;
+		if (cp_transaction) {
+			JBUFFER_TRACE(jh, "remove from old cp transaction");
+			__journal_remove_checkpoint(jh);
+		}
+
+		/* Only re-checkpoint the buffer_head if it is marked
+		 * dirty.  If the buffer was added to the BJ_Forget list
+		 * by journal_forget, it may no longer be dirty and
+		 * there's no point in keeping a checkpoint record for
+		 * it. */
+
+		/* A buffer which has been freed while still being
+		 * journaled by a previous transaction may end up still
+		 * being dirty here, but we want to avoid writing back
+		 * that buffer in the future now that the last use has
+		 * been committed.  That's not only a performance gain,
+		 * it also stops aliasing problems if the buffer is left
+		 * behind for writeback and gets reallocated for another
+		 * use in a different page. */
+		if (buffer_freed(bh)) {
+			clear_buffer_freed(bh);
+			clear_buffer_jbddirty(bh);
+		}
+
+		if (buffer_jbddirty(bh)) {
+			JBUFFER_TRACE(jh, "add to new checkpointing trans");
+			__journal_insert_checkpoint(jh, commit_transaction);
+			JBUFFER_TRACE(jh, "refile for checkpoint writeback");
+			__journal_refile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+		} else {
+			J_ASSERT_BH(bh, !buffer_dirty(bh));
+			J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+			__journal_unfile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+			journal_remove_journal_head(bh);  /* needs a brelse */
+			release_buffer_page(bh);
+		}
+		spin_unlock(&journal->j_list_lock);
+		if (cond_resched())
+			goto restart_loop;
+	}
+
+	/* Done with this transaction! */
+
+	jbd_debug(3, "JBD: commit phase 8\n");
+
+	J_ASSERT(commit_transaction->t_state == T_COMMIT);
+
+	/*
+	 * This is a bit sleazy.  We borrow j_list_lock to protect
+	 * journal->j_committing_transaction in __journal_remove_checkpoint.
+	 * Really, __jornal_remove_checkpoint should be using j_state_lock but
+	 * it's a bit hassle to hold that across __journal_remove_checkpoint
+	 */
+	spin_lock(&journal->j_state_lock);
+	spin_lock(&journal->j_list_lock);
+	commit_transaction->t_state = T_FINISHED;
+	J_ASSERT(commit_transaction == journal->j_committing_transaction);
+	journal->j_commit_sequence = commit_transaction->t_tid;
+	journal->j_committing_transaction = NULL;
+	spin_unlock(&journal->j_state_lock);
+
+	if (commit_transaction->t_checkpoint_list == NULL) {
+		__journal_drop_transaction(journal, commit_transaction);
+	} else {
+		if (journal->j_checkpoint_transactions == NULL) {
+			journal->j_checkpoint_transactions = commit_transaction;
+			commit_transaction->t_cpnext = commit_transaction;
+			commit_transaction->t_cpprev = commit_transaction;
+		} else {
+			commit_transaction->t_cpnext =
+				journal->j_checkpoint_transactions;
+			commit_transaction->t_cpprev =
+				commit_transaction->t_cpnext->t_cpprev;
+			commit_transaction->t_cpnext->t_cpprev =
+				commit_transaction;
+			commit_transaction->t_cpprev->t_cpnext =
+				commit_transaction;
+		}
+	}
+	spin_unlock(&journal->j_list_lock);
+
+	jbd_debug(1, "JBD: commit %d complete, head %d\n",
+		  journal->j_commit_sequence, journal->j_tail_sequence);
+
+	wake_up(&journal->j_wait_done_commit);
+}
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
new file mode 100644
index 00000000000..1e6f2e2ad4a
--- /dev/null
+++ b/fs/jbd/journal.c
@@ -0,0 +1,2003 @@
+/*
+ * linux/fs/journal.c
+ *
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
+ *
+ * Copyright 1998 Red Hat corp --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Generic filesystem journal-writing code; part of the ext2fs
+ * journaling system.
+ *
+ * This file manages journals: areas of disk reserved for logging
+ * transactional updates.  This includes the kernel journaling thread
+ * which is responsible for scheduling updates to the log.
+ *
+ * We do not actually manage the physical storage of the journal in this
+ * file: that is left to a per-journal policy function, which allows us
+ * to store the journal within a filesystem-specified area for ext2
+ * journaling (ext2 can use a reserved inode for storing the log).
+ */
+
+#include <linux/module.h>
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/smp_lock.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/suspend.h>
+#include <linux/pagemap.h>
+#include <asm/uaccess.h>
+#include <asm/page.h>
+#include <linux/proc_fs.h>
+
+EXPORT_SYMBOL(journal_start);
+EXPORT_SYMBOL(journal_restart);
+EXPORT_SYMBOL(journal_extend);
+EXPORT_SYMBOL(journal_stop);
+EXPORT_SYMBOL(journal_lock_updates);
+EXPORT_SYMBOL(journal_unlock_updates);
+EXPORT_SYMBOL(journal_get_write_access);
+EXPORT_SYMBOL(journal_get_create_access);
+EXPORT_SYMBOL(journal_get_undo_access);
+EXPORT_SYMBOL(journal_dirty_data);
+EXPORT_SYMBOL(journal_dirty_metadata);
+EXPORT_SYMBOL(journal_release_buffer);
+EXPORT_SYMBOL(journal_forget);
+#if 0
+EXPORT_SYMBOL(journal_sync_buffer);
+#endif
+EXPORT_SYMBOL(journal_flush);
+EXPORT_SYMBOL(journal_revoke);
+
+EXPORT_SYMBOL(journal_init_dev);
+EXPORT_SYMBOL(journal_init_inode);
+EXPORT_SYMBOL(journal_update_format);
+EXPORT_SYMBOL(journal_check_used_features);
+EXPORT_SYMBOL(journal_check_available_features);
+EXPORT_SYMBOL(journal_set_features);
+EXPORT_SYMBOL(journal_create);
+EXPORT_SYMBOL(journal_load);
+EXPORT_SYMBOL(journal_destroy);
+EXPORT_SYMBOL(journal_recover);
+EXPORT_SYMBOL(journal_update_superblock);
+EXPORT_SYMBOL(journal_abort);
+EXPORT_SYMBOL(journal_errno);
+EXPORT_SYMBOL(journal_ack_err);
+EXPORT_SYMBOL(journal_clear_err);
+EXPORT_SYMBOL(log_wait_commit);
+EXPORT_SYMBOL(journal_start_commit);
+EXPORT_SYMBOL(journal_force_commit_nested);
+EXPORT_SYMBOL(journal_wipe);
+EXPORT_SYMBOL(journal_blocks_per_page);
+EXPORT_SYMBOL(journal_invalidatepage);
+EXPORT_SYMBOL(journal_try_to_free_buffers);
+EXPORT_SYMBOL(journal_force_commit);
+
+static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
+
+/*
+ * Helper function used to manage commit timeouts
+ */
+
+static void commit_timeout(unsigned long __data)
+{
+	struct task_struct * p = (struct task_struct *) __data;
+
+	wake_up_process(p);
+}
+
+/* Static check for data structure consistency.  There's no code
+ * invoked --- we'll just get a linker failure if things aren't right.
+ */
+void __journal_internal_check(void)
+{
+	extern void journal_bad_superblock_size(void);
+	if (sizeof(struct journal_superblock_s) != 1024)
+		journal_bad_superblock_size();
+}
+
+/*
+ * kjournald: The main thread function used to manage a logging device
+ * journal.
+ *
+ * This kernel thread is responsible for two things:
+ *
+ * 1) COMMIT:  Every so often we need to commit the current state of the
+ *    filesystem to disk.  The journal thread is responsible for writing
+ *    all of the metadata buffers to disk.
+ *
+ * 2) CHECKPOINT: We cannot reuse a used section of the log file until all
+ *    of the data in that part of the log has been rewritten elsewhere on
+ *    the disk.  Flushing these old buffers to reclaim space in the log is
+ *    known as checkpointing, and this thread is responsible for that job.
+ */
+
+journal_t *current_journal;		// AKPM: debug
+
+int kjournald(void *arg)
+{
+	journal_t *journal = (journal_t *) arg;
+	transaction_t *transaction;
+	struct timer_list timer;
+
+	current_journal = journal;
+
+	daemonize("kjournald");
+
+	/* Set up an interval timer which can be used to trigger a
+           commit wakeup after the commit interval expires */
+	init_timer(&timer);
+	timer.data = (unsigned long) current;
+	timer.function = commit_timeout;
+	journal->j_commit_timer = &timer;
+
+	/* Record that the journal thread is running */
+	journal->j_task = current;
+	wake_up(&journal->j_wait_done_commit);
+
+	printk(KERN_INFO "kjournald starting.  Commit interval %ld seconds\n",
+			journal->j_commit_interval / HZ);
+
+	/*
+	 * And now, wait forever for commit wakeup events.
+	 */
+	spin_lock(&journal->j_state_lock);
+
+loop:
+	if (journal->j_flags & JFS_UNMOUNT)
+		goto end_loop;
+
+	jbd_debug(1, "commit_sequence=%d, commit_request=%d\n",
+		journal->j_commit_sequence, journal->j_commit_request);
+
+	if (journal->j_commit_sequence != journal->j_commit_request) {
+		jbd_debug(1, "OK, requests differ\n");
+		spin_unlock(&journal->j_state_lock);
+		del_timer_sync(journal->j_commit_timer);
+		journal_commit_transaction(journal);
+		spin_lock(&journal->j_state_lock);
+		goto loop;
+	}
+
+	wake_up(&journal->j_wait_done_commit);
+	if (current->flags & PF_FREEZE) {
+		/*
+		 * The simpler the better. Flushing journal isn't a
+		 * good idea, because that depends on threads that may
+		 * be already stopped.
+		 */
+		jbd_debug(1, "Now suspending kjournald\n");
+		spin_unlock(&journal->j_state_lock);
+		refrigerator(PF_FREEZE);
+		spin_lock(&journal->j_state_lock);
+	} else {
+		/*
+		 * We assume on resume that commits are already there,
+		 * so we don't sleep
+		 */
+		DEFINE_WAIT(wait);
+		int should_sleep = 1;
+
+		prepare_to_wait(&journal->j_wait_commit, &wait,
+				TASK_INTERRUPTIBLE);
+		if (journal->j_commit_sequence != journal->j_commit_request)
+			should_sleep = 0;
+		transaction = journal->j_running_transaction;
+		if (transaction && time_after_eq(jiffies,
+						transaction->t_expires))
+			should_sleep = 0;
+		if (should_sleep) {
+			spin_unlock(&journal->j_state_lock);
+			schedule();
+			spin_lock(&journal->j_state_lock);
+		}
+		finish_wait(&journal->j_wait_commit, &wait);
+	}
+
+	jbd_debug(1, "kjournald wakes\n");
+
+	/*
+	 * Were we woken up by a commit wakeup event?
+	 */
+	transaction = journal->j_running_transaction;
+	if (transaction && time_after_eq(jiffies, transaction->t_expires)) {
+		journal->j_commit_request = transaction->t_tid;
+		jbd_debug(1, "woke because of timeout\n");
+	}
+	goto loop;
+
+end_loop:
+	spin_unlock(&journal->j_state_lock);
+	del_timer_sync(journal->j_commit_timer);
+	journal->j_task = NULL;
+	wake_up(&journal->j_wait_done_commit);
+	jbd_debug(1, "Journal thread exiting.\n");
+	return 0;
+}
+
+static void journal_start_thread(journal_t *journal)
+{
+	kernel_thread(kjournald, journal, CLONE_VM|CLONE_FS|CLONE_FILES);
+	wait_event(journal->j_wait_done_commit, journal->j_task != 0);
+}
+
+static void journal_kill_thread(journal_t *journal)
+{
+	spin_lock(&journal->j_state_lock);
+	journal->j_flags |= JFS_UNMOUNT;
+
+	while (journal->j_task) {
+		wake_up(&journal->j_wait_commit);
+		spin_unlock(&journal->j_state_lock);
+		wait_event(journal->j_wait_done_commit, journal->j_task == 0);
+		spin_lock(&journal->j_state_lock);
+	}
+	spin_unlock(&journal->j_state_lock);
+}
+
+/*
+ * journal_write_metadata_buffer: write a metadata buffer to the journal.
+ *
+ * Writes a metadata buffer to a given disk block.  The actual IO is not
+ * performed but a new buffer_head is constructed which labels the data
+ * to be written with the correct destination disk block.
+ *
+ * Any magic-number escaping which needs to be done will cause a
+ * copy-out here.  If the buffer happens to start with the
+ * JFS_MAGIC_NUMBER, then we can't write it to the log directly: the
+ * magic number is only written to the log for descripter blocks.  In
+ * this case, we copy the data and replace the first word with 0, and we
+ * return a result code which indicates that this buffer needs to be
+ * marked as an escaped buffer in the corresponding log descriptor
+ * block.  The missing word can then be restored when the block is read
+ * during recovery.
+ *
+ * If the source buffer has already been modified by a new transaction
+ * since we took the last commit snapshot, we use the frozen copy of
+ * that data for IO.  If we end up using the existing buffer_head's data
+ * for the write, then we *have* to lock the buffer to prevent anyone
+ * else from using and possibly modifying it while the IO is in
+ * progress.
+ *
+ * The function returns a pointer to the buffer_heads to be used for IO.
+ *
+ * We assume that the journal has already been locked in this function.
+ *
+ * Return value:
+ *  <0: Error
+ * >=0: Finished OK
+ *
+ * On success:
+ * Bit 0 set == escape performed on the data
+ * Bit 1 set == buffer copy-out performed (kfree the data after IO)
+ */
+
+int journal_write_metadata_buffer(transaction_t *transaction,
+				  struct journal_head  *jh_in,
+				  struct journal_head **jh_out,
+				  int blocknr)
+{
+	int need_copy_out = 0;
+	int done_copy_out = 0;
+	int do_escape = 0;
+	char *mapped_data;
+	struct buffer_head *new_bh;
+	struct journal_head *new_jh;
+	struct page *new_page;
+	unsigned int new_offset;
+	struct buffer_head *bh_in = jh2bh(jh_in);
+
+	/*
+	 * The buffer really shouldn't be locked: only the current committing
+	 * transaction is allowed to write it, so nobody else is allowed
+	 * to do any IO.
+	 *
+	 * akpm: except if we're journalling data, and write() output is
+	 * also part of a shared mapping, and another thread has
+	 * decided to launch a writepage() against this buffer.
+	 */
+	J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in));
+
+	new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL);
+
+	/*
+	 * If a new transaction has already done a buffer copy-out, then
+	 * we use that version of the data for the commit.
+	 */
+	jbd_lock_bh_state(bh_in);
+repeat:
+	if (jh_in->b_frozen_data) {
+		done_copy_out = 1;
+		new_page = virt_to_page(jh_in->b_frozen_data);
+		new_offset = offset_in_page(jh_in->b_frozen_data);
+	} else {
+		new_page = jh2bh(jh_in)->b_page;
+		new_offset = offset_in_page(jh2bh(jh_in)->b_data);
+	}
+
+	mapped_data = kmap_atomic(new_page, KM_USER0);
+	/*
+	 * Check for escaping
+	 */
+	if (*((__be32 *)(mapped_data + new_offset)) ==
+				cpu_to_be32(JFS_MAGIC_NUMBER)) {
+		need_copy_out = 1;
+		do_escape = 1;
+	}
+	kunmap_atomic(mapped_data, KM_USER0);
+
+	/*
+	 * Do we need to do a data copy?
+	 */
+	if (need_copy_out && !done_copy_out) {
+		char *tmp;
+
+		jbd_unlock_bh_state(bh_in);
+		tmp = jbd_rep_kmalloc(bh_in->b_size, GFP_NOFS);
+		jbd_lock_bh_state(bh_in);
+		if (jh_in->b_frozen_data) {
+			kfree(tmp);
+			goto repeat;
+		}
+
+		jh_in->b_frozen_data = tmp;
+		mapped_data = kmap_atomic(new_page, KM_USER0);
+		memcpy(tmp, mapped_data + new_offset, jh2bh(jh_in)->b_size);
+		kunmap_atomic(mapped_data, KM_USER0);
+
+		new_page = virt_to_page(tmp);
+		new_offset = offset_in_page(tmp);
+		done_copy_out = 1;
+	}
+
+	/*
+	 * Did we need to do an escaping?  Now we've done all the
+	 * copying, we can finally do so.
+	 */
+	if (do_escape) {
+		mapped_data = kmap_atomic(new_page, KM_USER0);
+		*((unsigned int *)(mapped_data + new_offset)) = 0;
+		kunmap_atomic(mapped_data, KM_USER0);
+	}
+
+	/* keep subsequent assertions sane */
+	new_bh->b_state = 0;
+	init_buffer(new_bh, NULL, NULL);
+	atomic_set(&new_bh->b_count, 1);
+	jbd_unlock_bh_state(bh_in);
+
+	new_jh = journal_add_journal_head(new_bh);	/* This sleeps */
+
+	set_bh_page(new_bh, new_page, new_offset);
+	new_jh->b_transaction = NULL;
+	new_bh->b_size = jh2bh(jh_in)->b_size;
+	new_bh->b_bdev = transaction->t_journal->j_dev;
+	new_bh->b_blocknr = blocknr;
+	set_buffer_mapped(new_bh);
+	set_buffer_dirty(new_bh);
+
+	*jh_out = new_jh;
+
+	/*
+	 * The to-be-written buffer needs to get moved to the io queue,
+	 * and the original buffer whose contents we are shadowing or
+	 * copying is moved to the transaction's shadow queue.
+	 */
+	JBUFFER_TRACE(jh_in, "file as BJ_Shadow");
+	journal_file_buffer(jh_in, transaction, BJ_Shadow);
+	JBUFFER_TRACE(new_jh, "file as BJ_IO");
+	journal_file_buffer(new_jh, transaction, BJ_IO);
+
+	return do_escape | (done_copy_out << 1);
+}
+
+/*
+ * Allocation code for the journal file.  Manage the space left in the
+ * journal, so that we can begin checkpointing when appropriate.
+ */
+
+/*
+ * __log_space_left: Return the number of free blocks left in the journal.
+ *
+ * Called with the journal already locked.
+ *
+ * Called under j_state_lock
+ */
+
+int __log_space_left(journal_t *journal)
+{
+	int left = journal->j_free;
+
+	assert_spin_locked(&journal->j_state_lock);
+
+	/*
+	 * Be pessimistic here about the number of those free blocks which
+	 * might be required for log descriptor control blocks.
+	 */
+
+#define MIN_LOG_RESERVED_BLOCKS 32 /* Allow for rounding errors */
+
+	left -= MIN_LOG_RESERVED_BLOCKS;
+
+	if (left <= 0)
+		return 0;
+	left -= (left >> 3);
+	return left;
+}
+
+/*
+ * Called under j_state_lock.  Returns true if a transaction was started.
+ */
+int __log_start_commit(journal_t *journal, tid_t target)
+{
+	/*
+	 * Are we already doing a recent enough commit?
+	 */
+	if (!tid_geq(journal->j_commit_request, target)) {
+		/*
+		 * We want a new commit: OK, mark the request and wakup the
+		 * commit thread.  We do _not_ do the commit ourselves.
+		 */
+
+		journal->j_commit_request = target;
+		jbd_debug(1, "JBD: requesting commit %d/%d\n",
+			  journal->j_commit_request,
+			  journal->j_commit_sequence);
+		wake_up(&journal->j_wait_commit);
+		return 1;
+	}
+	return 0;
+}
+
+int log_start_commit(journal_t *journal, tid_t tid)
+{
+	int ret;
+
+	spin_lock(&journal->j_state_lock);
+	ret = __log_start_commit(journal, tid);
+	spin_unlock(&journal->j_state_lock);
+	return ret;
+}
+
+/*
+ * Force and wait upon a commit if the calling process is not within
+ * transaction.  This is used for forcing out undo-protected data which contains
+ * bitmaps, when the fs is running out of space.
+ *
+ * We can only force the running transaction if we don't have an active handle;
+ * otherwise, we will deadlock.
+ *
+ * Returns true if a transaction was started.
+ */
+int journal_force_commit_nested(journal_t *journal)
+{
+	transaction_t *transaction = NULL;
+	tid_t tid;
+
+	spin_lock(&journal->j_state_lock);
+	if (journal->j_running_transaction && !current->journal_info) {
+		transaction = journal->j_running_transaction;
+		__log_start_commit(journal, transaction->t_tid);
+	} else if (journal->j_committing_transaction)
+		transaction = journal->j_committing_transaction;
+
+	if (!transaction) {
+		spin_unlock(&journal->j_state_lock);
+		return 0;	/* Nothing to retry */
+	}
+
+	tid = transaction->t_tid;
+	spin_unlock(&journal->j_state_lock);
+	log_wait_commit(journal, tid);
+	return 1;
+}
+
+/*
+ * Start a commit of the current running transaction (if any).  Returns true
+ * if a transaction was started, and fills its tid in at *ptid
+ */
+int journal_start_commit(journal_t *journal, tid_t *ptid)
+{
+	int ret = 0;
+
+	spin_lock(&journal->j_state_lock);
+	if (journal->j_running_transaction) {
+		tid_t tid = journal->j_running_transaction->t_tid;
+
+		ret = __log_start_commit(journal, tid);
+		if (ret && ptid)
+			*ptid = tid;
+	} else if (journal->j_committing_transaction && ptid) {
+		/*
+		 * If ext3_write_super() recently started a commit, then we
+		 * have to wait for completion of that transaction
+		 */
+		*ptid = journal->j_committing_transaction->t_tid;
+		ret = 1;
+	}
+	spin_unlock(&journal->j_state_lock);
+	return ret;
+}
+
+/*
+ * Wait for a specified commit to complete.
+ * The caller may not hold the journal lock.
+ */
+int log_wait_commit(journal_t *journal, tid_t tid)
+{
+	int err = 0;
+
+#ifdef CONFIG_JBD_DEBUG
+	spin_lock(&journal->j_state_lock);
+	if (!tid_geq(journal->j_commit_request, tid)) {
+		printk(KERN_EMERG
+		       "%s: error: j_commit_request=%d, tid=%d\n",
+		       __FUNCTION__, journal->j_commit_request, tid);
+	}
+	spin_unlock(&journal->j_state_lock);
+#endif
+	spin_lock(&journal->j_state_lock);
+	while (tid_gt(tid, journal->j_commit_sequence)) {
+		jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n",
+				  tid, journal->j_commit_sequence);
+		wake_up(&journal->j_wait_commit);
+		spin_unlock(&journal->j_state_lock);
+		wait_event(journal->j_wait_done_commit,
+				!tid_gt(tid, journal->j_commit_sequence));
+		spin_lock(&journal->j_state_lock);
+	}
+	spin_unlock(&journal->j_state_lock);
+
+	if (unlikely(is_journal_aborted(journal))) {
+		printk(KERN_EMERG "journal commit I/O error\n");
+		err = -EIO;
+	}
+	return err;
+}
+
+/*
+ * Log buffer allocation routines:
+ */
+
+int journal_next_log_block(journal_t *journal, unsigned long *retp)
+{
+	unsigned long blocknr;
+
+	spin_lock(&journal->j_state_lock);
+	J_ASSERT(journal->j_free > 1);
+
+	blocknr = journal->j_head;
+	journal->j_head++;
+	journal->j_free--;
+	if (journal->j_head == journal->j_last)
+		journal->j_head = journal->j_first;
+	spin_unlock(&journal->j_state_lock);
+	return journal_bmap(journal, blocknr, retp);
+}
+
+/*
+ * Conversion of logical to physical block numbers for the journal
+ *
+ * On external journals the journal blocks are identity-mapped, so
+ * this is a no-op.  If needed, we can use j_blk_offset - everything is
+ * ready.
+ */
+int journal_bmap(journal_t *journal, unsigned long blocknr, 
+		 unsigned long *retp)
+{
+	int err = 0;
+	unsigned long ret;
+
+	if (journal->j_inode) {
+		ret = bmap(journal->j_inode, blocknr);
+		if (ret)
+			*retp = ret;
+		else {
+			char b[BDEVNAME_SIZE];
+
+			printk(KERN_ALERT "%s: journal block not found "
+					"at offset %lu on %s\n",
+				__FUNCTION__,
+				blocknr,
+				bdevname(journal->j_dev, b));
+			err = -EIO;
+			__journal_abort_soft(journal, err);
+		}
+	} else {
+		*retp = blocknr; /* +journal->j_blk_offset */
+	}
+	return err;
+}
+
+/*
+ * We play buffer_head aliasing tricks to write data/metadata blocks to
+ * the journal without copying their contents, but for journal
+ * descriptor blocks we do need to generate bona fide buffers.
+ *
+ * After the caller of journal_get_descriptor_buffer() has finished modifying
+ * the buffer's contents they really should run flush_dcache_page(bh->b_page).
+ * But we don't bother doing that, so there will be coherency problems with
+ * mmaps of blockdevs which hold live JBD-controlled filesystems.
+ */
+struct journal_head *journal_get_descriptor_buffer(journal_t *journal)
+{
+	struct buffer_head *bh;
+	unsigned long blocknr;
+	int err;
+
+	err = journal_next_log_block(journal, &blocknr);
+
+	if (err)
+		return NULL;
+
+	bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
+	lock_buffer(bh);
+	memset(bh->b_data, 0, journal->j_blocksize);
+	set_buffer_uptodate(bh);
+	unlock_buffer(bh);
+	BUFFER_TRACE(bh, "return this buffer");
+	return journal_add_journal_head(bh);
+}
+
+/*
+ * Management for journal control blocks: functions to create and
+ * destroy journal_t structures, and to initialise and read existing
+ * journal blocks from disk.  */
+
+/* First: create and setup a journal_t object in memory.  We initialise
+ * very few fields yet: that has to wait until we have created the
+ * journal structures from from scratch, or loaded them from disk. */
+
+static journal_t * journal_init_common (void)
+{
+	journal_t *journal;
+	int err;
+
+	journal = jbd_kmalloc(sizeof(*journal), GFP_KERNEL);
+	if (!journal)
+		goto fail;
+	memset(journal, 0, sizeof(*journal));
+
+	init_waitqueue_head(&journal->j_wait_transaction_locked);
+	init_waitqueue_head(&journal->j_wait_logspace);
+	init_waitqueue_head(&journal->j_wait_done_commit);
+	init_waitqueue_head(&journal->j_wait_checkpoint);
+	init_waitqueue_head(&journal->j_wait_commit);
+	init_waitqueue_head(&journal->j_wait_updates);
+	init_MUTEX(&journal->j_barrier);
+	init_MUTEX(&journal->j_checkpoint_sem);
+	spin_lock_init(&journal->j_revoke_lock);
+	spin_lock_init(&journal->j_list_lock);
+	spin_lock_init(&journal->j_state_lock);
+
+	journal->j_commit_interval = (HZ * JBD_DEFAULT_MAX_COMMIT_AGE);
+
+	/* The journal is marked for error until we succeed with recovery! */
+	journal->j_flags = JFS_ABORT;
+
+	/* Set up a default-sized revoke table for the new mount. */
+	err = journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH);
+	if (err) {
+		kfree(journal);
+		goto fail;
+	}
+	return journal;
+fail:
+	return NULL;
+}
+
+/* journal_init_dev and journal_init_inode:
+ *
+ * Create a journal structure assigned some fixed set of disk blocks to
+ * the journal.  We don't actually touch those disk blocks yet, but we
+ * need to set up all of the mapping information to tell the journaling
+ * system where the journal blocks are.
+ *
+ */
+
+/**
+ *  journal_t * journal_init_dev() - creates an initialises a journal structure
+ *  @bdev: Block device on which to create the journal
+ *  @fs_dev: Device which hold journalled filesystem for this journal.
+ *  @start: Block nr Start of journal.
+ *  @len:  Lenght of the journal in blocks.
+ *  @blocksize: blocksize of journalling device
+ *  @returns: a newly created journal_t *
+ *  
+ *  journal_init_dev creates a journal which maps a fixed contiguous
+ *  range of blocks on an arbitrary block device.
+ * 
+ */
+journal_t * journal_init_dev(struct block_device *bdev,
+			struct block_device *fs_dev,
+			int start, int len, int blocksize)
+{
+	journal_t *journal = journal_init_common();
+	struct buffer_head *bh;
+	int n;
+
+	if (!journal)
+		return NULL;
+
+	journal->j_dev = bdev;
+	journal->j_fs_dev = fs_dev;
+	journal->j_blk_offset = start;
+	journal->j_maxlen = len;
+	journal->j_blocksize = blocksize;
+
+	bh = __getblk(journal->j_dev, start, journal->j_blocksize);
+	J_ASSERT(bh != NULL);
+	journal->j_sb_buffer = bh;
+	journal->j_superblock = (journal_superblock_t *)bh->b_data;
+
+	/* journal descriptor can store up to n blocks -bzzz */
+	n = journal->j_blocksize / sizeof(journal_block_tag_t);
+	journal->j_wbufsize = n;
+	journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
+	if (!journal->j_wbuf) {
+		printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
+			__FUNCTION__);
+		kfree(journal);
+		journal = NULL;
+	}
+
+	return journal;
+}
+ 
+/** 
+ *  journal_t * journal_init_inode () - creates a journal which maps to a inode.
+ *  @inode: An inode to create the journal in
+ *  
+ * journal_init_inode creates a journal which maps an on-disk inode as
+ * the journal.  The inode must exist already, must support bmap() and
+ * must have all data blocks preallocated.
+ */
+journal_t * journal_init_inode (struct inode *inode)
+{
+	struct buffer_head *bh;
+	journal_t *journal = journal_init_common();
+	int err;
+	int n;
+	unsigned long blocknr;
+
+	if (!journal)
+		return NULL;
+
+	journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev;
+	journal->j_inode = inode;
+	jbd_debug(1,
+		  "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n",
+		  journal, inode->i_sb->s_id, inode->i_ino, 
+		  (long long) inode->i_size,
+		  inode->i_sb->s_blocksize_bits, inode->i_sb->s_blocksize);
+
+	journal->j_maxlen = inode->i_size >> inode->i_sb->s_blocksize_bits;
+	journal->j_blocksize = inode->i_sb->s_blocksize;
+
+	/* journal descriptor can store up to n blocks -bzzz */
+	n = journal->j_blocksize / sizeof(journal_block_tag_t);
+	journal->j_wbufsize = n;
+	journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
+	if (!journal->j_wbuf) {
+		printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
+			__FUNCTION__);
+		kfree(journal);
+		return NULL;
+	}
+
+	err = journal_bmap(journal, 0, &blocknr);
+	/* If that failed, give up */
+	if (err) {
+		printk(KERN_ERR "%s: Cannnot locate journal superblock\n",
+		       __FUNCTION__);
+		kfree(journal);
+		return NULL;
+	}
+
+	bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
+	J_ASSERT(bh != NULL);
+	journal->j_sb_buffer = bh;
+	journal->j_superblock = (journal_superblock_t *)bh->b_data;
+
+	return journal;
+}
+
+/* 
+ * If the journal init or create aborts, we need to mark the journal
+ * superblock as being NULL to prevent the journal destroy from writing
+ * back a bogus superblock. 
+ */
+static void journal_fail_superblock (journal_t *journal)
+{
+	struct buffer_head *bh = journal->j_sb_buffer;
+	brelse(bh);
+	journal->j_sb_buffer = NULL;
+}
+
+/*
+ * Given a journal_t structure, initialise the various fields for
+ * startup of a new journaling session.  We use this both when creating
+ * a journal, and after recovering an old journal to reset it for
+ * subsequent use.
+ */
+
+static int journal_reset(journal_t *journal)
+{
+	journal_superblock_t *sb = journal->j_superblock;
+	unsigned int first, last;
+
+	first = be32_to_cpu(sb->s_first);
+	last = be32_to_cpu(sb->s_maxlen);
+
+	journal->j_first = first;
+	journal->j_last = last;
+
+	journal->j_head = first;
+	journal->j_tail = first;
+	journal->j_free = last - first;
+
+	journal->j_tail_sequence = journal->j_transaction_sequence;
+	journal->j_commit_sequence = journal->j_transaction_sequence - 1;
+	journal->j_commit_request = journal->j_commit_sequence;
+
+	journal->j_max_transaction_buffers = journal->j_maxlen / 4;
+
+	/* Add the dynamic fields and write it to disk. */
+	journal_update_superblock(journal, 1);
+	journal_start_thread(journal);
+	return 0;
+}
+
+/** 
+ * int journal_create() - Initialise the new journal file
+ * @journal: Journal to create. This structure must have been initialised
+ * 
+ * Given a journal_t structure which tells us which disk blocks we can
+ * use, create a new journal superblock and initialise all of the
+ * journal fields from scratch.  
+ **/
+int journal_create(journal_t *journal)
+{
+	unsigned long blocknr;
+	struct buffer_head *bh;
+	journal_superblock_t *sb;
+	int i, err;
+
+	if (journal->j_maxlen < JFS_MIN_JOURNAL_BLOCKS) {
+		printk (KERN_ERR "Journal length (%d blocks) too short.\n",
+			journal->j_maxlen);
+		journal_fail_superblock(journal);
+		return -EINVAL;
+	}
+
+	if (journal->j_inode == NULL) {
+		/*
+		 * We don't know what block to start at!
+		 */
+		printk(KERN_EMERG
+		       "%s: creation of journal on external device!\n",
+		       __FUNCTION__);
+		BUG();
+	}
+
+	/* Zero out the entire journal on disk.  We cannot afford to
+	   have any blocks on disk beginning with JFS_MAGIC_NUMBER. */
+	jbd_debug(1, "JBD: Zeroing out journal blocks...\n");
+	for (i = 0; i < journal->j_maxlen; i++) {
+		err = journal_bmap(journal, i, &blocknr);
+		if (err)
+			return err;
+		bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
+		lock_buffer(bh);
+		memset (bh->b_data, 0, journal->j_blocksize);
+		BUFFER_TRACE(bh, "marking dirty");
+		mark_buffer_dirty(bh);
+		BUFFER_TRACE(bh, "marking uptodate");
+		set_buffer_uptodate(bh);
+		unlock_buffer(bh);
+		__brelse(bh);
+	}
+
+	sync_blockdev(journal->j_dev);
+	jbd_debug(1, "JBD: journal cleared.\n");
+
+	/* OK, fill in the initial static fields in the new superblock */
+	sb = journal->j_superblock;
+
+	sb->s_header.h_magic	 = cpu_to_be32(JFS_MAGIC_NUMBER);
+	sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2);
+
+	sb->s_blocksize	= cpu_to_be32(journal->j_blocksize);
+	sb->s_maxlen	= cpu_to_be32(journal->j_maxlen);
+	sb->s_first	= cpu_to_be32(1);
+
+	journal->j_transaction_sequence = 1;
+
+	journal->j_flags &= ~JFS_ABORT;
+	journal->j_format_version = 2;
+
+	return journal_reset(journal);
+}
+
+/** 
+ * void journal_update_superblock() - Update journal sb on disk.
+ * @journal: The journal to update.
+ * @wait: Set to '0' if you don't want to wait for IO completion.
+ *
+ * Update a journal's dynamic superblock fields and write it to disk,
+ * optionally waiting for the IO to complete.
+ */
+void journal_update_superblock(journal_t *journal, int wait)
+{
+	journal_superblock_t *sb = journal->j_superblock;
+	struct buffer_head *bh = journal->j_sb_buffer;
+
+	/*
+	 * As a special case, if the on-disk copy is already marked as needing
+	 * no recovery (s_start == 0) and there are no outstanding transactions
+	 * in the filesystem, then we can safely defer the superblock update
+	 * until the next commit by setting JFS_FLUSHED.  This avoids
+	 * attempting a write to a potential-readonly device.
+	 */
+	if (sb->s_start == 0 && journal->j_tail_sequence ==
+				journal->j_transaction_sequence) {
+		jbd_debug(1,"JBD: Skipping superblock update on recovered sb "
+			"(start %ld, seq %d, errno %d)\n",
+			journal->j_tail, journal->j_tail_sequence, 
+			journal->j_errno);
+		goto out;
+	}
+
+	spin_lock(&journal->j_state_lock);
+	jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n",
+		  journal->j_tail, journal->j_tail_sequence, journal->j_errno);
+
+	sb->s_sequence = cpu_to_be32(journal->j_tail_sequence);
+	sb->s_start    = cpu_to_be32(journal->j_tail);
+	sb->s_errno    = cpu_to_be32(journal->j_errno);
+	spin_unlock(&journal->j_state_lock);
+
+	BUFFER_TRACE(bh, "marking dirty");
+	mark_buffer_dirty(bh);
+	if (wait)
+		sync_dirty_buffer(bh);
+	else
+		ll_rw_block(WRITE, 1, &bh);
+
+out:
+	/* If we have just flushed the log (by marking s_start==0), then
+	 * any future commit will have to be careful to update the
+	 * superblock again to re-record the true start of the log. */
+
+	spin_lock(&journal->j_state_lock);
+	if (sb->s_start)
+		journal->j_flags &= ~JFS_FLUSHED;
+	else
+		journal->j_flags |= JFS_FLUSHED;
+	spin_unlock(&journal->j_state_lock);
+}
+
+/*
+ * Read the superblock for a given journal, performing initial
+ * validation of the format.
+ */
+
+static int journal_get_superblock(journal_t *journal)
+{
+	struct buffer_head *bh;
+	journal_superblock_t *sb;
+	int err = -EIO;
+
+	bh = journal->j_sb_buffer;
+
+	J_ASSERT(bh != NULL);
+	if (!buffer_uptodate(bh)) {
+		ll_rw_block(READ, 1, &bh);
+		wait_on_buffer(bh);
+		if (!buffer_uptodate(bh)) {
+			printk (KERN_ERR
+				"JBD: IO error reading journal superblock\n");
+			goto out;
+		}
+	}
+
+	sb = journal->j_superblock;
+
+	err = -EINVAL;
+
+	if (sb->s_header.h_magic != cpu_to_be32(JFS_MAGIC_NUMBER) ||
+	    sb->s_blocksize != cpu_to_be32(journal->j_blocksize)) {
+		printk(KERN_WARNING "JBD: no valid journal superblock found\n");
+		goto out;
+	}
+
+	switch(be32_to_cpu(sb->s_header.h_blocktype)) {
+	case JFS_SUPERBLOCK_V1:
+		journal->j_format_version = 1;
+		break;
+	case JFS_SUPERBLOCK_V2:
+		journal->j_format_version = 2;
+		break;
+	default:
+		printk(KERN_WARNING "JBD: unrecognised superblock format ID\n");
+		goto out;
+	}
+
+	if (be32_to_cpu(sb->s_maxlen) < journal->j_maxlen)
+		journal->j_maxlen = be32_to_cpu(sb->s_maxlen);
+	else if (be32_to_cpu(sb->s_maxlen) > journal->j_maxlen) {
+		printk (KERN_WARNING "JBD: journal file too short\n");
+		goto out;
+	}
+
+	return 0;
+
+out:
+	journal_fail_superblock(journal);
+	return err;
+}
+
+/*
+ * Load the on-disk journal superblock and read the key fields into the
+ * journal_t.
+ */
+
+static int load_superblock(journal_t *journal)
+{
+	int err;
+	journal_superblock_t *sb;
+
+	err = journal_get_superblock(journal);
+	if (err)
+		return err;
+
+	sb = journal->j_superblock;
+
+	journal->j_tail_sequence = be32_to_cpu(sb->s_sequence);
+	journal->j_tail = be32_to_cpu(sb->s_start);
+	journal->j_first = be32_to_cpu(sb->s_first);
+	journal->j_last = be32_to_cpu(sb->s_maxlen);
+	journal->j_errno = be32_to_cpu(sb->s_errno);
+
+	return 0;
+}
+
+
+/**
+ * int journal_load() - Read journal from disk.
+ * @journal: Journal to act on.
+ * 
+ * Given a journal_t structure which tells us which disk blocks contain
+ * a journal, read the journal from disk to initialise the in-memory
+ * structures.
+ */
+int journal_load(journal_t *journal)
+{
+	int err;
+
+	err = load_superblock(journal);
+	if (err)
+		return err;
+
+	/* If this is a V2 superblock, then we have to check the
+	 * features flags on it. */
+
+	if (journal->j_format_version >= 2) {
+		journal_superblock_t *sb = journal->j_superblock;
+
+		if ((sb->s_feature_ro_compat &
+		     ~cpu_to_be32(JFS_KNOWN_ROCOMPAT_FEATURES)) ||
+		    (sb->s_feature_incompat &
+		     ~cpu_to_be32(JFS_KNOWN_INCOMPAT_FEATURES))) {
+			printk (KERN_WARNING
+				"JBD: Unrecognised features on journal\n");
+			return -EINVAL;
+		}
+	}
+
+	/* Let the recovery code check whether it needs to recover any
+	 * data from the journal. */
+	if (journal_recover(journal))
+		goto recovery_error;
+
+	/* OK, we've finished with the dynamic journal bits:
+	 * reinitialise the dynamic contents of the superblock in memory
+	 * and reset them on disk. */
+	if (journal_reset(journal))
+		goto recovery_error;
+
+	journal->j_flags &= ~JFS_ABORT;
+	journal->j_flags |= JFS_LOADED;
+	return 0;
+
+recovery_error:
+	printk (KERN_WARNING "JBD: recovery failed\n");
+	return -EIO;
+}
+
+/**
+ * void journal_destroy() - Release a journal_t structure.
+ * @journal: Journal to act on.
+ *
+ * Release a journal_t structure once it is no longer in use by the
+ * journaled object.
+ */
+void journal_destroy(journal_t *journal)
+{
+	/* Wait for the commit thread to wake up and die. */
+	journal_kill_thread(journal);
+
+	/* Force a final log commit */
+	if (journal->j_running_transaction)
+		journal_commit_transaction(journal);
+
+	/* Force any old transactions to disk */
+
+	/* Totally anal locking here... */
+	spin_lock(&journal->j_list_lock);
+	while (journal->j_checkpoint_transactions != NULL) {
+		spin_unlock(&journal->j_list_lock);
+		log_do_checkpoint(journal);
+		spin_lock(&journal->j_list_lock);
+	}
+
+	J_ASSERT(journal->j_running_transaction == NULL);
+	J_ASSERT(journal->j_committing_transaction == NULL);
+	J_ASSERT(journal->j_checkpoint_transactions == NULL);
+	spin_unlock(&journal->j_list_lock);
+
+	/* We can now mark the journal as empty. */
+	journal->j_tail = 0;
+	journal->j_tail_sequence = ++journal->j_transaction_sequence;
+	if (journal->j_sb_buffer) {
+		journal_update_superblock(journal, 1);
+		brelse(journal->j_sb_buffer);
+	}
+
+	if (journal->j_inode)
+		iput(journal->j_inode);
+	if (journal->j_revoke)
+		journal_destroy_revoke(journal);
+	kfree(journal->j_wbuf);
+	kfree(journal);
+}
+
+
+/**
+ *int journal_check_used_features () - Check if features specified are used.
+ * @journal: Journal to check.
+ * @compat: bitmask of compatible features
+ * @ro: bitmask of features that force read-only mount
+ * @incompat: bitmask of incompatible features
+ * 
+ * Check whether the journal uses all of a given set of
+ * features.  Return true (non-zero) if it does. 
+ **/
+
+int journal_check_used_features (journal_t *journal, unsigned long compat,
+				 unsigned long ro, unsigned long incompat)
+{
+	journal_superblock_t *sb;
+
+	if (!compat && !ro && !incompat)
+		return 1;
+	if (journal->j_format_version == 1)
+		return 0;
+
+	sb = journal->j_superblock;
+
+	if (((be32_to_cpu(sb->s_feature_compat) & compat) == compat) &&
+	    ((be32_to_cpu(sb->s_feature_ro_compat) & ro) == ro) &&
+	    ((be32_to_cpu(sb->s_feature_incompat) & incompat) == incompat))
+		return 1;
+
+	return 0;
+}
+
+/**
+ * int journal_check_available_features() - Check feature set in journalling layer
+ * @journal: Journal to check.
+ * @compat: bitmask of compatible features
+ * @ro: bitmask of features that force read-only mount
+ * @incompat: bitmask of incompatible features
+ * 
+ * Check whether the journaling code supports the use of
+ * all of a given set of features on this journal.  Return true
+ * (non-zero) if it can. */
+
+int journal_check_available_features (journal_t *journal, unsigned long compat,
+				      unsigned long ro, unsigned long incompat)
+{
+	journal_superblock_t *sb;
+
+	if (!compat && !ro && !incompat)
+		return 1;
+
+	sb = journal->j_superblock;
+
+	/* We can support any known requested features iff the
+	 * superblock is in version 2.  Otherwise we fail to support any
+	 * extended sb features. */
+
+	if (journal->j_format_version != 2)
+		return 0;
+
+	if ((compat   & JFS_KNOWN_COMPAT_FEATURES) == compat &&
+	    (ro       & JFS_KNOWN_ROCOMPAT_FEATURES) == ro &&
+	    (incompat & JFS_KNOWN_INCOMPAT_FEATURES) == incompat)
+		return 1;
+
+	return 0;
+}
+
+/**
+ * int journal_set_features () - Mark a given journal feature in the superblock
+ * @journal: Journal to act on.
+ * @compat: bitmask of compatible features
+ * @ro: bitmask of features that force read-only mount
+ * @incompat: bitmask of incompatible features
+ *
+ * Mark a given journal feature as present on the
+ * superblock.  Returns true if the requested features could be set. 
+ *
+ */
+
+int journal_set_features (journal_t *journal, unsigned long compat,
+			  unsigned long ro, unsigned long incompat)
+{
+	journal_superblock_t *sb;
+
+	if (journal_check_used_features(journal, compat, ro, incompat))
+		return 1;
+
+	if (!journal_check_available_features(journal, compat, ro, incompat))
+		return 0;
+
+	jbd_debug(1, "Setting new features 0x%lx/0x%lx/0x%lx\n",
+		  compat, ro, incompat);
+
+	sb = journal->j_superblock;
+
+	sb->s_feature_compat    |= cpu_to_be32(compat);
+	sb->s_feature_ro_compat |= cpu_to_be32(ro);
+	sb->s_feature_incompat  |= cpu_to_be32(incompat);
+
+	return 1;
+}
+
+
+/**
+ * int journal_update_format () - Update on-disk journal structure.
+ * @journal: Journal to act on.
+ *
+ * Given an initialised but unloaded journal struct, poke about in the
+ * on-disk structure to update it to the most recent supported version.
+ */
+int journal_update_format (journal_t *journal)
+{
+	journal_superblock_t *sb;
+	int err;
+
+	err = journal_get_superblock(journal);
+	if (err)
+		return err;
+
+	sb = journal->j_superblock;
+
+	switch (be32_to_cpu(sb->s_header.h_blocktype)) {
+	case JFS_SUPERBLOCK_V2:
+		return 0;
+	case JFS_SUPERBLOCK_V1:
+		return journal_convert_superblock_v1(journal, sb);
+	default:
+		break;
+	}
+	return -EINVAL;
+}
+
+static int journal_convert_superblock_v1(journal_t *journal,
+					 journal_superblock_t *sb)
+{
+	int offset, blocksize;
+	struct buffer_head *bh;
+
+	printk(KERN_WARNING
+		"JBD: Converting superblock from version 1 to 2.\n");
+
+	/* Pre-initialise new fields to zero */
+	offset = ((char *) &(sb->s_feature_compat)) - ((char *) sb);
+	blocksize = be32_to_cpu(sb->s_blocksize);
+	memset(&sb->s_feature_compat, 0, blocksize-offset);
+
+	sb->s_nr_users = cpu_to_be32(1);
+	sb->s_header.h_blocktype = cpu_to_be32(JFS_SUPERBLOCK_V2);
+	journal->j_format_version = 2;
+
+	bh = journal->j_sb_buffer;
+	BUFFER_TRACE(bh, "marking dirty");
+	mark_buffer_dirty(bh);
+	sync_dirty_buffer(bh);
+	return 0;
+}
+
+
+/**
+ * int journal_flush () - Flush journal
+ * @journal: Journal to act on.
+ * 
+ * Flush all data for a given journal to disk and empty the journal.
+ * Filesystems can use this when remounting readonly to ensure that
+ * recovery does not need to happen on remount.
+ */
+
+int journal_flush(journal_t *journal)
+{
+	int err = 0;
+	transaction_t *transaction = NULL;
+	unsigned long old_tail;
+
+	spin_lock(&journal->j_state_lock);
+
+	/* Force everything buffered to the log... */
+	if (journal->j_running_transaction) {
+		transaction = journal->j_running_transaction;
+		__log_start_commit(journal, transaction->t_tid);
+	} else if (journal->j_committing_transaction)
+		transaction = journal->j_committing_transaction;
+
+	/* Wait for the log commit to complete... */
+	if (transaction) {
+		tid_t tid = transaction->t_tid;
+
+		spin_unlock(&journal->j_state_lock);
+		log_wait_commit(journal, tid);
+	} else {
+		spin_unlock(&journal->j_state_lock);
+	}
+
+	/* ...and flush everything in the log out to disk. */
+	spin_lock(&journal->j_list_lock);
+	while (!err && journal->j_checkpoint_transactions != NULL) {
+		spin_unlock(&journal->j_list_lock);
+		err = log_do_checkpoint(journal);
+		spin_lock(&journal->j_list_lock);
+	}
+	spin_unlock(&journal->j_list_lock);
+	cleanup_journal_tail(journal);
+
+	/* Finally, mark the journal as really needing no recovery.
+	 * This sets s_start==0 in the underlying superblock, which is
+	 * the magic code for a fully-recovered superblock.  Any future
+	 * commits of data to the journal will restore the current
+	 * s_start value. */
+	spin_lock(&journal->j_state_lock);
+	old_tail = journal->j_tail;
+	journal->j_tail = 0;
+	spin_unlock(&journal->j_state_lock);
+	journal_update_superblock(journal, 1);
+	spin_lock(&journal->j_state_lock);
+	journal->j_tail = old_tail;
+
+	J_ASSERT(!journal->j_running_transaction);
+	J_ASSERT(!journal->j_committing_transaction);
+	J_ASSERT(!journal->j_checkpoint_transactions);
+	J_ASSERT(journal->j_head == journal->j_tail);
+	J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
+	spin_unlock(&journal->j_state_lock);
+	return err;
+}
+
+/**
+ * int journal_wipe() - Wipe journal contents
+ * @journal: Journal to act on.
+ * @write: flag (see below)
+ * 
+ * Wipe out all of the contents of a journal, safely.  This will produce
+ * a warning if the journal contains any valid recovery information.
+ * Must be called between journal_init_*() and journal_load().
+ *
+ * If 'write' is non-zero, then we wipe out the journal on disk; otherwise
+ * we merely suppress recovery.
+ */
+
+int journal_wipe(journal_t *journal, int write)
+{
+	journal_superblock_t *sb;
+	int err = 0;
+
+	J_ASSERT (!(journal->j_flags & JFS_LOADED));
+
+	err = load_superblock(journal);
+	if (err)
+		return err;
+
+	sb = journal->j_superblock;
+
+	if (!journal->j_tail)
+		goto no_recovery;
+
+	printk (KERN_WARNING "JBD: %s recovery information on journal\n",
+		write ? "Clearing" : "Ignoring");
+
+	err = journal_skip_recovery(journal);
+	if (write)
+		journal_update_superblock(journal, 1);
+
+ no_recovery:
+	return err;
+}
+
+/*
+ * journal_dev_name: format a character string to describe on what
+ * device this journal is present.
+ */
+
+const char *journal_dev_name(journal_t *journal, char *buffer)
+{
+	struct block_device *bdev;
+
+	if (journal->j_inode)
+		bdev = journal->j_inode->i_sb->s_bdev;
+	else
+		bdev = journal->j_dev;
+
+	return bdevname(bdev, buffer);
+}
+
+/*
+ * Journal abort has very specific semantics, which we describe
+ * for journal abort. 
+ *
+ * Two internal function, which provide abort to te jbd layer
+ * itself are here.
+ */
+
+/*
+ * Quick version for internal journal use (doesn't lock the journal).
+ * Aborts hard --- we mark the abort as occurred, but do _nothing_ else,
+ * and don't attempt to make any other journal updates.
+ */
+void __journal_abort_hard(journal_t *journal)
+{
+	transaction_t *transaction;
+	char b[BDEVNAME_SIZE];
+
+	if (journal->j_flags & JFS_ABORT)
+		return;
+
+	printk(KERN_ERR "Aborting journal on device %s.\n",
+		journal_dev_name(journal, b));
+
+	spin_lock(&journal->j_state_lock);
+	journal->j_flags |= JFS_ABORT;
+	transaction = journal->j_running_transaction;
+	if (transaction)
+		__log_start_commit(journal, transaction->t_tid);
+	spin_unlock(&journal->j_state_lock);
+}
+
+/* Soft abort: record the abort error status in the journal superblock,
+ * but don't do any other IO. */
+void __journal_abort_soft (journal_t *journal, int errno)
+{
+	if (journal->j_flags & JFS_ABORT)
+		return;
+
+	if (!journal->j_errno)
+		journal->j_errno = errno;
+
+	__journal_abort_hard(journal);
+
+	if (errno)
+		journal_update_superblock(journal, 1);
+}
+
+/**
+ * void journal_abort () - Shutdown the journal immediately.
+ * @journal: the journal to shutdown.
+ * @errno:   an error number to record in the journal indicating
+ *           the reason for the shutdown.
+ *
+ * Perform a complete, immediate shutdown of the ENTIRE
+ * journal (not of a single transaction).  This operation cannot be
+ * undone without closing and reopening the journal.
+ *           
+ * The journal_abort function is intended to support higher level error
+ * recovery mechanisms such as the ext2/ext3 remount-readonly error
+ * mode.
+ *
+ * Journal abort has very specific semantics.  Any existing dirty,
+ * unjournaled buffers in the main filesystem will still be written to
+ * disk by bdflush, but the journaling mechanism will be suspended
+ * immediately and no further transaction commits will be honoured.
+ *
+ * Any dirty, journaled buffers will be written back to disk without
+ * hitting the journal.  Atomicity cannot be guaranteed on an aborted
+ * filesystem, but we _do_ attempt to leave as much data as possible
+ * behind for fsck to use for cleanup.
+ *
+ * Any attempt to get a new transaction handle on a journal which is in
+ * ABORT state will just result in an -EROFS error return.  A
+ * journal_stop on an existing handle will return -EIO if we have
+ * entered abort state during the update.
+ *
+ * Recursive transactions are not disturbed by journal abort until the
+ * final journal_stop, which will receive the -EIO error.
+ *
+ * Finally, the journal_abort call allows the caller to supply an errno
+ * which will be recorded (if possible) in the journal superblock.  This
+ * allows a client to record failure conditions in the middle of a
+ * transaction without having to complete the transaction to record the
+ * failure to disk.  ext3_error, for example, now uses this
+ * functionality.
+ *
+ * Errors which originate from within the journaling layer will NOT
+ * supply an errno; a null errno implies that absolutely no further
+ * writes are done to the journal (unless there are any already in
+ * progress).
+ * 
+ */
+
+void journal_abort(journal_t *journal, int errno)
+{
+	__journal_abort_soft(journal, errno);
+}
+
+/** 
+ * int journal_errno () - returns the journal's error state.
+ * @journal: journal to examine.
+ *
+ * This is the errno numbet set with journal_abort(), the last
+ * time the journal was mounted - if the journal was stopped
+ * without calling abort this will be 0.
+ *
+ * If the journal has been aborted on this mount time -EROFS will
+ * be returned.
+ */
+int journal_errno(journal_t *journal)
+{
+	int err;
+
+	spin_lock(&journal->j_state_lock);
+	if (journal->j_flags & JFS_ABORT)
+		err = -EROFS;
+	else
+		err = journal->j_errno;
+	spin_unlock(&journal->j_state_lock);
+	return err;
+}
+
+/** 
+ * int journal_clear_err () - clears the journal's error state
+ * @journal: journal to act on.
+ *
+ * An error must be cleared or Acked to take a FS out of readonly
+ * mode.
+ */
+int journal_clear_err(journal_t *journal)
+{
+	int err = 0;
+
+	spin_lock(&journal->j_state_lock);
+	if (journal->j_flags & JFS_ABORT)
+		err = -EROFS;
+	else
+		journal->j_errno = 0;
+	spin_unlock(&journal->j_state_lock);
+	return err;
+}
+
+/** 
+ * void journal_ack_err() - Ack journal err.
+ * @journal: journal to act on.
+ *
+ * An error must be cleared or Acked to take a FS out of readonly
+ * mode.
+ */
+void journal_ack_err(journal_t *journal)
+{
+	spin_lock(&journal->j_state_lock);
+	if (journal->j_errno)
+		journal->j_flags |= JFS_ACK_ERR;
+	spin_unlock(&journal->j_state_lock);
+}
+
+int journal_blocks_per_page(struct inode *inode)
+{
+	return 1 << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits);
+}
+
+/*
+ * Simple support for retrying memory allocations.  Introduced to help to
+ * debug different VM deadlock avoidance strategies. 
+ */
+void * __jbd_kmalloc (const char *where, size_t size, int flags, int retry)
+{
+	return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0));
+}
+
+/*
+ * Journal_head storage management
+ */
+static kmem_cache_t *journal_head_cache;
+#ifdef CONFIG_JBD_DEBUG
+static atomic_t nr_journal_heads = ATOMIC_INIT(0);
+#endif
+
+static int journal_init_journal_head_cache(void)
+{
+	int retval;
+
+	J_ASSERT(journal_head_cache == 0);
+	journal_head_cache = kmem_cache_create("journal_head",
+				sizeof(struct journal_head),
+				0,		/* offset */
+				0,		/* flags */
+				NULL,		/* ctor */
+				NULL);		/* dtor */
+	retval = 0;
+	if (journal_head_cache == 0) {
+		retval = -ENOMEM;
+		printk(KERN_EMERG "JBD: no memory for journal_head cache\n");
+	}
+	return retval;
+}
+
+static void journal_destroy_journal_head_cache(void)
+{
+	J_ASSERT(journal_head_cache != NULL);
+	kmem_cache_destroy(journal_head_cache);
+	journal_head_cache = NULL;
+}
+
+/*
+ * journal_head splicing and dicing
+ */
+static struct journal_head *journal_alloc_journal_head(void)
+{
+	struct journal_head *ret;
+	static unsigned long last_warning;
+
+#ifdef CONFIG_JBD_DEBUG
+	atomic_inc(&nr_journal_heads);
+#endif
+	ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS);
+	if (ret == 0) {
+		jbd_debug(1, "out of memory for journal_head\n");
+		if (time_after(jiffies, last_warning + 5*HZ)) {
+			printk(KERN_NOTICE "ENOMEM in %s, retrying.\n",
+			       __FUNCTION__);
+			last_warning = jiffies;
+		}
+		while (ret == 0) {
+			yield();
+			ret = kmem_cache_alloc(journal_head_cache, GFP_NOFS);
+		}
+	}
+	return ret;
+}
+
+static void journal_free_journal_head(struct journal_head *jh)
+{
+#ifdef CONFIG_JBD_DEBUG
+	atomic_dec(&nr_journal_heads);
+	memset(jh, 0x5b, sizeof(*jh));
+#endif
+	kmem_cache_free(journal_head_cache, jh);
+}
+
+/*
+ * A journal_head is attached to a buffer_head whenever JBD has an
+ * interest in the buffer.
+ *
+ * Whenever a buffer has an attached journal_head, its ->b_state:BH_JBD bit
+ * is set.  This bit is tested in core kernel code where we need to take
+ * JBD-specific actions.  Testing the zeroness of ->b_private is not reliable
+ * there.
+ *
+ * When a buffer has its BH_JBD bit set, its ->b_count is elevated by one.
+ *
+ * When a buffer has its BH_JBD bit set it is immune from being released by
+ * core kernel code, mainly via ->b_count.
+ *
+ * A journal_head may be detached from its buffer_head when the journal_head's
+ * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL.
+ * Various places in JBD call journal_remove_journal_head() to indicate that the
+ * journal_head can be dropped if needed.
+ *
+ * Various places in the kernel want to attach a journal_head to a buffer_head
+ * _before_ attaching the journal_head to a transaction.  To protect the
+ * journal_head in this situation, journal_add_journal_head elevates the
+ * journal_head's b_jcount refcount by one.  The caller must call
+ * journal_put_journal_head() to undo this.
+ *
+ * So the typical usage would be:
+ *
+ *	(Attach a journal_head if needed.  Increments b_jcount)
+ *	struct journal_head *jh = journal_add_journal_head(bh);
+ *	...
+ *	jh->b_transaction = xxx;
+ *	journal_put_journal_head(jh);
+ *
+ * Now, the journal_head's b_jcount is zero, but it is safe from being released
+ * because it has a non-zero b_transaction.
+ */
+
+/*
+ * Give a buffer_head a journal_head.
+ *
+ * Doesn't need the journal lock.
+ * May sleep.
+ */
+struct journal_head *journal_add_journal_head(struct buffer_head *bh)
+{
+	struct journal_head *jh;
+	struct journal_head *new_jh = NULL;
+
+repeat:
+	if (!buffer_jbd(bh)) {
+		new_jh = journal_alloc_journal_head();
+		memset(new_jh, 0, sizeof(*new_jh));
+	}
+
+	jbd_lock_bh_journal_head(bh);
+	if (buffer_jbd(bh)) {
+		jh = bh2jh(bh);
+	} else {
+		J_ASSERT_BH(bh,
+			(atomic_read(&bh->b_count) > 0) ||
+			(bh->b_page && bh->b_page->mapping));
+
+		if (!new_jh) {
+			jbd_unlock_bh_journal_head(bh);
+			goto repeat;
+		}
+
+		jh = new_jh;
+		new_jh = NULL;		/* We consumed it */
+		set_buffer_jbd(bh);
+		bh->b_private = jh;
+		jh->b_bh = bh;
+		get_bh(bh);
+		BUFFER_TRACE(bh, "added journal_head");
+	}
+	jh->b_jcount++;
+	jbd_unlock_bh_journal_head(bh);
+	if (new_jh)
+		journal_free_journal_head(new_jh);
+	return bh->b_private;
+}
+
+/*
+ * Grab a ref against this buffer_head's journal_head.  If it ended up not
+ * having a journal_head, return NULL
+ */
+struct journal_head *journal_grab_journal_head(struct buffer_head *bh)
+{
+	struct journal_head *jh = NULL;
+
+	jbd_lock_bh_journal_head(bh);
+	if (buffer_jbd(bh)) {
+		jh = bh2jh(bh);
+		jh->b_jcount++;
+	}
+	jbd_unlock_bh_journal_head(bh);
+	return jh;
+}
+
+static void __journal_remove_journal_head(struct buffer_head *bh)
+{
+	struct journal_head *jh = bh2jh(bh);
+
+	J_ASSERT_JH(jh, jh->b_jcount >= 0);
+
+	get_bh(bh);
+	if (jh->b_jcount == 0) {
+		if (jh->b_transaction == NULL &&
+				jh->b_next_transaction == NULL &&
+				jh->b_cp_transaction == NULL) {
+			J_ASSERT_JH(jh, jh->b_jlist == BJ_None);
+			J_ASSERT_BH(bh, buffer_jbd(bh));
+			J_ASSERT_BH(bh, jh2bh(jh) == bh);
+			BUFFER_TRACE(bh, "remove journal_head");
+			if (jh->b_frozen_data) {
+				printk(KERN_WARNING "%s: freeing "
+						"b_frozen_data\n",
+						__FUNCTION__);
+				kfree(jh->b_frozen_data);
+			}
+			if (jh->b_committed_data) {
+				printk(KERN_WARNING "%s: freeing "
+						"b_committed_data\n",
+						__FUNCTION__);
+				kfree(jh->b_committed_data);
+			}
+			bh->b_private = NULL;
+			jh->b_bh = NULL;	/* debug, really */
+			clear_buffer_jbd(bh);
+			__brelse(bh);
+			journal_free_journal_head(jh);
+		} else {
+			BUFFER_TRACE(bh, "journal_head was locked");
+		}
+	}
+}
+
+/*
+ * journal_remove_journal_head(): if the buffer isn't attached to a transaction
+ * and has a zero b_jcount then remove and release its journal_head.   If we did
+ * see that the buffer is not used by any transaction we also "logically"
+ * decrement ->b_count.
+ *
+ * We in fact take an additional increment on ->b_count as a convenience,
+ * because the caller usually wants to do additional things with the bh
+ * after calling here.
+ * The caller of journal_remove_journal_head() *must* run __brelse(bh) at some
+ * time.  Once the caller has run __brelse(), the buffer is eligible for
+ * reaping by try_to_free_buffers().
+ */
+void journal_remove_journal_head(struct buffer_head *bh)
+{
+	jbd_lock_bh_journal_head(bh);
+	__journal_remove_journal_head(bh);
+	jbd_unlock_bh_journal_head(bh);
+}
+
+/*
+ * Drop a reference on the passed journal_head.  If it fell to zero then try to
+ * release the journal_head from the buffer_head.
+ */
+void journal_put_journal_head(struct journal_head *jh)
+{
+	struct buffer_head *bh = jh2bh(jh);
+
+	jbd_lock_bh_journal_head(bh);
+	J_ASSERT_JH(jh, jh->b_jcount > 0);
+	--jh->b_jcount;
+	if (!jh->b_jcount && !jh->b_transaction) {
+		__journal_remove_journal_head(bh);
+		__brelse(bh);
+	}
+	jbd_unlock_bh_journal_head(bh);
+}
+
+/*
+ * /proc tunables
+ */
+#if defined(CONFIG_JBD_DEBUG)
+int journal_enable_debug;
+EXPORT_SYMBOL(journal_enable_debug);
+#endif
+
+#if defined(CONFIG_JBD_DEBUG) && defined(CONFIG_PROC_FS)
+
+static struct proc_dir_entry *proc_jbd_debug;
+
+int read_jbd_debug(char *page, char **start, off_t off,
+			  int count, int *eof, void *data)
+{
+	int ret;
+
+	ret = sprintf(page + off, "%d\n", journal_enable_debug);
+	*eof = 1;
+	return ret;
+}
+
+int write_jbd_debug(struct file *file, const char __user *buffer,
+			   unsigned long count, void *data)
+{
+	char buf[32];
+
+	if (count > ARRAY_SIZE(buf) - 1)
+		count = ARRAY_SIZE(buf) - 1;
+	if (copy_from_user(buf, buffer, count))
+		return -EFAULT;
+	buf[ARRAY_SIZE(buf) - 1] = '\0';
+	journal_enable_debug = simple_strtoul(buf, NULL, 10);
+	return count;
+}
+
+#define JBD_PROC_NAME "sys/fs/jbd-debug"
+
+static void __init create_jbd_proc_entry(void)
+{
+	proc_jbd_debug = create_proc_entry(JBD_PROC_NAME, 0644, NULL);
+	if (proc_jbd_debug) {
+		/* Why is this so hard? */
+		proc_jbd_debug->read_proc = read_jbd_debug;
+		proc_jbd_debug->write_proc = write_jbd_debug;
+	}
+}
+
+static void __exit remove_jbd_proc_entry(void)
+{
+	if (proc_jbd_debug)
+		remove_proc_entry(JBD_PROC_NAME, NULL);
+}
+
+#else
+
+#define create_jbd_proc_entry() do {} while (0)
+#define remove_jbd_proc_entry() do {} while (0)
+
+#endif
+
+kmem_cache_t *jbd_handle_cache;
+
+static int __init journal_init_handle_cache(void)
+{
+	jbd_handle_cache = kmem_cache_create("journal_handle",
+				sizeof(handle_t),
+				0,		/* offset */
+				0,		/* flags */
+				NULL,		/* ctor */
+				NULL);		/* dtor */
+	if (jbd_handle_cache == NULL) {
+		printk(KERN_EMERG "JBD: failed to create handle cache\n");
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+static void journal_destroy_handle_cache(void)
+{
+	if (jbd_handle_cache)
+		kmem_cache_destroy(jbd_handle_cache);
+}
+
+/*
+ * Module startup and shutdown
+ */
+
+static int __init journal_init_caches(void)
+{
+	int ret;
+
+	ret = journal_init_revoke_caches();
+	if (ret == 0)
+		ret = journal_init_journal_head_cache();
+	if (ret == 0)
+		ret = journal_init_handle_cache();
+	return ret;
+}
+
+static void journal_destroy_caches(void)
+{
+	journal_destroy_revoke_caches();
+	journal_destroy_journal_head_cache();
+	journal_destroy_handle_cache();
+}
+
+static int __init journal_init(void)
+{
+	int ret;
+
+	ret = journal_init_caches();
+	if (ret != 0)
+		journal_destroy_caches();
+	create_jbd_proc_entry();
+	return ret;
+}
+
+static void __exit journal_exit(void)
+{
+#ifdef CONFIG_JBD_DEBUG
+	int n = atomic_read(&nr_journal_heads);
+	if (n)
+		printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n);
+#endif
+	remove_jbd_proc_entry();
+	journal_destroy_caches();
+}
+
+MODULE_LICENSE("GPL");
+module_init(journal_init);
+module_exit(journal_exit);
+
diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
new file mode 100644
index 00000000000..103c34e4fb2
--- /dev/null
+++ b/fs/jbd/recovery.c
@@ -0,0 +1,591 @@
+/*
+ * linux/fs/recovery.c
+ * 
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
+ *
+ * Copyright 1999-2000 Red Hat Software --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Journal recovery routines for the generic filesystem journaling code;
+ * part of the ext2fs journaling system.  
+ */
+
+#ifndef __KERNEL__
+#include "jfs_user.h"
+#else
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#endif
+
+/*
+ * Maintain information about the progress of the recovery job, so that
+ * the different passes can carry information between them. 
+ */
+struct recovery_info 
+{
+	tid_t		start_transaction;
+	tid_t		end_transaction;
+
+	int		nr_replays;
+	int		nr_revokes;
+	int		nr_revoke_hits;
+};
+
+enum passtype {PASS_SCAN, PASS_REVOKE, PASS_REPLAY};
+static int do_one_pass(journal_t *journal,
+				struct recovery_info *info, enum passtype pass);
+static int scan_revoke_records(journal_t *, struct buffer_head *,
+				tid_t, struct recovery_info *);
+
+#ifdef __KERNEL__
+
+/* Release readahead buffers after use */
+void journal_brelse_array(struct buffer_head *b[], int n)
+{
+	while (--n >= 0)
+		brelse (b[n]);
+}
+
+
+/*
+ * When reading from the journal, we are going through the block device
+ * layer directly and so there is no readahead being done for us.  We
+ * need to implement any readahead ourselves if we want it to happen at
+ * all.  Recovery is basically one long sequential read, so make sure we
+ * do the IO in reasonably large chunks.
+ *
+ * This is not so critical that we need to be enormously clever about
+ * the readahead size, though.  128K is a purely arbitrary, good-enough
+ * fixed value.
+ */
+
+#define MAXBUF 8
+static int do_readahead(journal_t *journal, unsigned int start)
+{
+	int err;
+	unsigned int max, nbufs, next;
+	unsigned long blocknr;
+	struct buffer_head *bh;
+
+	struct buffer_head * bufs[MAXBUF];
+
+	/* Do up to 128K of readahead */
+	max = start + (128 * 1024 / journal->j_blocksize);
+	if (max > journal->j_maxlen)
+		max = journal->j_maxlen;
+
+	/* Do the readahead itself.  We'll submit MAXBUF buffer_heads at
+	 * a time to the block device IO layer. */
+
+	nbufs = 0;
+
+	for (next = start; next < max; next++) {
+		err = journal_bmap(journal, next, &blocknr);
+
+		if (err) {
+			printk (KERN_ERR "JBD: bad block at offset %u\n",
+				next);
+			goto failed;
+		}
+
+		bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
+		if (!bh) {
+			err = -ENOMEM;
+			goto failed;
+		}
+
+		if (!buffer_uptodate(bh) && !buffer_locked(bh)) {
+			bufs[nbufs++] = bh;
+			if (nbufs == MAXBUF) {
+				ll_rw_block(READ, nbufs, bufs);
+				journal_brelse_array(bufs, nbufs);
+				nbufs = 0;
+			}
+		} else
+			brelse(bh);
+	}
+
+	if (nbufs)
+		ll_rw_block(READ, nbufs, bufs);
+	err = 0;
+
+failed:
+	if (nbufs) 
+		journal_brelse_array(bufs, nbufs);
+	return err;
+}
+
+#endif /* __KERNEL__ */
+
+
+/*
+ * Read a block from the journal
+ */
+
+static int jread(struct buffer_head **bhp, journal_t *journal, 
+		 unsigned int offset)
+{
+	int err;
+	unsigned long blocknr;
+	struct buffer_head *bh;
+
+	*bhp = NULL;
+
+	if (offset >= journal->j_maxlen) {
+		printk(KERN_ERR "JBD: corrupted journal superblock\n");
+		return -EIO;
+	}
+
+	err = journal_bmap(journal, offset, &blocknr);
+
+	if (err) {
+		printk (KERN_ERR "JBD: bad block at offset %u\n",
+			offset);
+		return err;
+	}
+
+	bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize);
+	if (!bh)
+		return -ENOMEM;
+
+	if (!buffer_uptodate(bh)) {
+		/* If this is a brand new buffer, start readahead.
+                   Otherwise, we assume we are already reading it.  */
+		if (!buffer_req(bh))
+			do_readahead(journal, offset);
+		wait_on_buffer(bh);
+	}
+
+	if (!buffer_uptodate(bh)) {
+		printk (KERN_ERR "JBD: Failed to read block at offset %u\n",
+			offset);
+		brelse(bh);
+		return -EIO;
+	}
+
+	*bhp = bh;
+	return 0;
+}
+
+
+/*
+ * Count the number of in-use tags in a journal descriptor block.
+ */
+
+static int count_tags(struct buffer_head *bh, int size)
+{
+	char *			tagp;
+	journal_block_tag_t *	tag;
+	int			nr = 0;
+
+	tagp = &bh->b_data[sizeof(journal_header_t)];
+
+	while ((tagp - bh->b_data + sizeof(journal_block_tag_t)) <= size) {
+		tag = (journal_block_tag_t *) tagp;
+
+		nr++;
+		tagp += sizeof(journal_block_tag_t);
+		if (!(tag->t_flags & cpu_to_be32(JFS_FLAG_SAME_UUID)))
+			tagp += 16;
+
+		if (tag->t_flags & cpu_to_be32(JFS_FLAG_LAST_TAG))
+			break;
+	}
+
+	return nr;
+}
+
+
+/* Make sure we wrap around the log correctly! */
+#define wrap(journal, var)						\
+do {									\
+	if (var >= (journal)->j_last)					\
+		var -= ((journal)->j_last - (journal)->j_first);	\
+} while (0)
+
+/**
+ * int journal_recover(journal_t *journal) - recovers a on-disk journal
+ * @journal: the journal to recover
+ * 
+ * The primary function for recovering the log contents when mounting a
+ * journaled device.  
+ *
+ * Recovery is done in three passes.  In the first pass, we look for the
+ * end of the log.  In the second, we assemble the list of revoke
+ * blocks.  In the third and final pass, we replay any un-revoked blocks
+ * in the log.  
+ */
+int journal_recover(journal_t *journal)
+{
+	int			err;
+	journal_superblock_t *	sb;
+
+	struct recovery_info	info;
+
+	memset(&info, 0, sizeof(info));
+	sb = journal->j_superblock;
+
+	/* 
+	 * The journal superblock's s_start field (the current log head)
+	 * is always zero if, and only if, the journal was cleanly
+	 * unmounted.  
+	 */
+
+	if (!sb->s_start) {
+		jbd_debug(1, "No recovery required, last transaction %d\n",
+			  be32_to_cpu(sb->s_sequence));
+		journal->j_transaction_sequence = be32_to_cpu(sb->s_sequence) + 1;
+		return 0;
+	}
+
+	err = do_one_pass(journal, &info, PASS_SCAN);
+	if (!err)
+		err = do_one_pass(journal, &info, PASS_REVOKE);
+	if (!err)
+		err = do_one_pass(journal, &info, PASS_REPLAY);
+
+	jbd_debug(0, "JBD: recovery, exit status %d, "
+		  "recovered transactions %u to %u\n",
+		  err, info.start_transaction, info.end_transaction);
+	jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", 
+		  info.nr_replays, info.nr_revoke_hits, info.nr_revokes);
+
+	/* Restart the log at the next transaction ID, thus invalidating
+	 * any existing commit records in the log. */
+	journal->j_transaction_sequence = ++info.end_transaction;
+
+	journal_clear_revoke(journal);
+	sync_blockdev(journal->j_fs_dev);
+	return err;
+}
+
+/**
+ * int journal_skip_recovery() - Start journal and wipe exiting records 
+ * @journal: journal to startup
+ * 
+ * Locate any valid recovery information from the journal and set up the
+ * journal structures in memory to ignore it (presumably because the
+ * caller has evidence that it is out of date).  
+ * This function does'nt appear to be exorted..
+ *
+ * We perform one pass over the journal to allow us to tell the user how
+ * much recovery information is being erased, and to let us initialise
+ * the journal transaction sequence numbers to the next unused ID. 
+ */
+int journal_skip_recovery(journal_t *journal)
+{
+	int			err;
+	journal_superblock_t *	sb;
+
+	struct recovery_info	info;
+
+	memset (&info, 0, sizeof(info));
+	sb = journal->j_superblock;
+
+	err = do_one_pass(journal, &info, PASS_SCAN);
+
+	if (err) {
+		printk(KERN_ERR "JBD: error %d scanning journal\n", err);
+		++journal->j_transaction_sequence;
+	} else {
+#ifdef CONFIG_JBD_DEBUG
+		int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence);
+#endif
+		jbd_debug(0, 
+			  "JBD: ignoring %d transaction%s from the journal.\n",
+			  dropped, (dropped == 1) ? "" : "s");
+		journal->j_transaction_sequence = ++info.end_transaction;
+	}
+
+	journal->j_tail = 0;
+	return err;
+}
+
+static int do_one_pass(journal_t *journal,
+			struct recovery_info *info, enum passtype pass)
+{
+	unsigned int		first_commit_ID, next_commit_ID;
+	unsigned long		next_log_block;
+	int			err, success = 0;
+	journal_superblock_t *	sb;
+	journal_header_t * 	tmp;
+	struct buffer_head *	bh;
+	unsigned int		sequence;
+	int			blocktype;
+
+	/* Precompute the maximum metadata descriptors in a descriptor block */
+	int			MAX_BLOCKS_PER_DESC;
+	MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t))
+			       / sizeof(journal_block_tag_t));
+
+	/* 
+	 * First thing is to establish what we expect to find in the log
+	 * (in terms of transaction IDs), and where (in terms of log
+	 * block offsets): query the superblock.  
+	 */
+
+	sb = journal->j_superblock;
+	next_commit_ID = be32_to_cpu(sb->s_sequence);
+	next_log_block = be32_to_cpu(sb->s_start);
+
+	first_commit_ID = next_commit_ID;
+	if (pass == PASS_SCAN)
+		info->start_transaction = first_commit_ID;
+
+	jbd_debug(1, "Starting recovery pass %d\n", pass);
+
+	/*
+	 * Now we walk through the log, transaction by transaction,
+	 * making sure that each transaction has a commit block in the
+	 * expected place.  Each complete transaction gets replayed back
+	 * into the main filesystem. 
+	 */
+
+	while (1) {
+		int			flags;
+		char *			tagp;
+		journal_block_tag_t *	tag;
+		struct buffer_head *	obh;
+		struct buffer_head *	nbh;
+
+		cond_resched();		/* We're under lock_kernel() */
+
+		/* If we already know where to stop the log traversal,
+		 * check right now that we haven't gone past the end of
+		 * the log. */
+
+		if (pass != PASS_SCAN)
+			if (tid_geq(next_commit_ID, info->end_transaction))
+				break;
+
+		jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n",
+			  next_commit_ID, next_log_block, journal->j_last);
+
+		/* Skip over each chunk of the transaction looking
+		 * either the next descriptor block or the final commit
+		 * record. */
+
+		jbd_debug(3, "JBD: checking block %ld\n", next_log_block);
+		err = jread(&bh, journal, next_log_block);
+		if (err)
+			goto failed;
+
+		next_log_block++;
+		wrap(journal, next_log_block);
+
+		/* What kind of buffer is it? 
+		 * 
+		 * If it is a descriptor block, check that it has the
+		 * expected sequence number.  Otherwise, we're all done
+		 * here. */
+
+		tmp = (journal_header_t *)bh->b_data;
+
+		if (tmp->h_magic != cpu_to_be32(JFS_MAGIC_NUMBER)) {
+			brelse(bh);
+			break;
+		}
+
+		blocktype = be32_to_cpu(tmp->h_blocktype);
+		sequence = be32_to_cpu(tmp->h_sequence);
+		jbd_debug(3, "Found magic %d, sequence %d\n", 
+			  blocktype, sequence);
+
+		if (sequence != next_commit_ID) {
+			brelse(bh);
+			break;
+		}
+
+		/* OK, we have a valid descriptor block which matches
+		 * all of the sequence number checks.  What are we going
+		 * to do with it?  That depends on the pass... */
+
+		switch(blocktype) {
+		case JFS_DESCRIPTOR_BLOCK:
+			/* If it is a valid descriptor block, replay it
+			 * in pass REPLAY; otherwise, just skip over the
+			 * blocks it describes. */
+			if (pass != PASS_REPLAY) {
+				next_log_block +=
+					count_tags(bh, journal->j_blocksize);
+				wrap(journal, next_log_block);
+				brelse(bh);
+				continue;
+			}
+
+			/* A descriptor block: we can now write all of
+			 * the data blocks.  Yay, useful work is finally
+			 * getting done here! */
+
+			tagp = &bh->b_data[sizeof(journal_header_t)];
+			while ((tagp - bh->b_data +sizeof(journal_block_tag_t))
+			       <= journal->j_blocksize) {
+				unsigned long io_block;
+
+				tag = (journal_block_tag_t *) tagp;
+				flags = be32_to_cpu(tag->t_flags);
+
+				io_block = next_log_block++;
+				wrap(journal, next_log_block);
+				err = jread(&obh, journal, io_block);
+				if (err) {
+					/* Recover what we can, but
+					 * report failure at the end. */
+					success = err;
+					printk (KERN_ERR 
+						"JBD: IO error %d recovering "
+						"block %ld in log\n",
+						err, io_block);
+				} else {
+					unsigned long blocknr;
+
+					J_ASSERT(obh != NULL);
+					blocknr = be32_to_cpu(tag->t_blocknr);
+
+					/* If the block has been
+					 * revoked, then we're all done
+					 * here. */
+					if (journal_test_revoke
+					    (journal, blocknr, 
+					     next_commit_ID)) {
+						brelse(obh);
+						++info->nr_revoke_hits;
+						goto skip_write;
+					}
+
+					/* Find a buffer for the new
+					 * data being restored */
+					nbh = __getblk(journal->j_fs_dev,
+							blocknr,
+							journal->j_blocksize);
+					if (nbh == NULL) {
+						printk(KERN_ERR 
+						       "JBD: Out of memory "
+						       "during recovery.\n");
+						err = -ENOMEM;
+						brelse(bh);
+						brelse(obh);
+						goto failed;
+					}
+
+					lock_buffer(nbh);
+					memcpy(nbh->b_data, obh->b_data,
+							journal->j_blocksize);
+					if (flags & JFS_FLAG_ESCAPE) {
+						*((__be32 *)bh->b_data) =
+						cpu_to_be32(JFS_MAGIC_NUMBER);
+					}
+
+					BUFFER_TRACE(nbh, "marking dirty");
+					set_buffer_uptodate(nbh);
+					mark_buffer_dirty(nbh);
+					BUFFER_TRACE(nbh, "marking uptodate");
+					++info->nr_replays;
+					/* ll_rw_block(WRITE, 1, &nbh); */
+					unlock_buffer(nbh);
+					brelse(obh);
+					brelse(nbh);
+				}
+
+			skip_write:
+				tagp += sizeof(journal_block_tag_t);
+				if (!(flags & JFS_FLAG_SAME_UUID))
+					tagp += 16;
+
+				if (flags & JFS_FLAG_LAST_TAG)
+					break;
+			}
+
+			brelse(bh);
+			continue;
+
+		case JFS_COMMIT_BLOCK:
+			/* Found an expected commit block: not much to
+			 * do other than move on to the next sequence
+			 * number. */
+			brelse(bh);
+			next_commit_ID++;
+			continue;
+
+		case JFS_REVOKE_BLOCK:
+			/* If we aren't in the REVOKE pass, then we can
+			 * just skip over this block. */
+			if (pass != PASS_REVOKE) {
+				brelse(bh);
+				continue;
+			}
+
+			err = scan_revoke_records(journal, bh,
+						  next_commit_ID, info);
+			brelse(bh);
+			if (err)
+				goto failed;
+			continue;
+
+		default:
+			jbd_debug(3, "Unrecognised magic %d, end of scan.\n",
+				  blocktype);
+			goto done;
+		}
+	}
+
+ done:
+	/* 
+	 * We broke out of the log scan loop: either we came to the
+	 * known end of the log or we found an unexpected block in the
+	 * log.  If the latter happened, then we know that the "current"
+	 * transaction marks the end of the valid log.
+	 */
+
+	if (pass == PASS_SCAN)
+		info->end_transaction = next_commit_ID;
+	else {
+		/* It's really bad news if different passes end up at
+		 * different places (but possible due to IO errors). */
+		if (info->end_transaction != next_commit_ID) {
+			printk (KERN_ERR "JBD: recovery pass %d ended at "
+				"transaction %u, expected %u\n",
+				pass, next_commit_ID, info->end_transaction);
+			if (!success)
+				success = -EIO;
+		}
+	}
+
+	return success;
+
+ failed:
+	return err;
+}
+
+
+/* Scan a revoke record, marking all blocks mentioned as revoked. */
+
+static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, 
+			       tid_t sequence, struct recovery_info *info)
+{
+	journal_revoke_header_t *header;
+	int offset, max;
+
+	header = (journal_revoke_header_t *) bh->b_data;
+	offset = sizeof(journal_revoke_header_t);
+	max = be32_to_cpu(header->r_count);
+
+	while (offset < max) {
+		unsigned long blocknr;
+		int err;
+
+		blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset)));
+		offset += 4;
+		err = journal_set_revoke(journal, blocknr, sequence);
+		if (err)
+			return err;
+		++info->nr_revokes;
+	}
+	return 0;
+}
diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
new file mode 100644
index 00000000000..d327a598f86
--- /dev/null
+++ b/fs/jbd/revoke.c
@@ -0,0 +1,702 @@
+/*
+ * linux/fs/revoke.c
+ * 
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 2000
+ *
+ * Copyright 2000 Red Hat corp --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Journal revoke routines for the generic filesystem journaling code;
+ * part of the ext2fs journaling system.
+ *
+ * Revoke is the mechanism used to prevent old log records for deleted
+ * metadata from being replayed on top of newer data using the same
+ * blocks.  The revoke mechanism is used in two separate places:
+ * 
+ * + Commit: during commit we write the entire list of the current
+ *   transaction's revoked blocks to the journal
+ * 
+ * + Recovery: during recovery we record the transaction ID of all
+ *   revoked blocks.  If there are multiple revoke records in the log
+ *   for a single block, only the last one counts, and if there is a log
+ *   entry for a block beyond the last revoke, then that log entry still
+ *   gets replayed.
+ *
+ * We can get interactions between revokes and new log data within a
+ * single transaction:
+ *
+ * Block is revoked and then journaled:
+ *   The desired end result is the journaling of the new block, so we 
+ *   cancel the revoke before the transaction commits.
+ *
+ * Block is journaled and then revoked:
+ *   The revoke must take precedence over the write of the block, so we
+ *   need either to cancel the journal entry or to write the revoke
+ *   later in the log than the log block.  In this case, we choose the
+ *   latter: journaling a block cancels any revoke record for that block
+ *   in the current transaction, so any revoke for that block in the
+ *   transaction must have happened after the block was journaled and so
+ *   the revoke must take precedence.
+ *
+ * Block is revoked and then written as data: 
+ *   The data write is allowed to succeed, but the revoke is _not_
+ *   cancelled.  We still need to prevent old log records from
+ *   overwriting the new data.  We don't even need to clear the revoke
+ *   bit here.
+ *
+ * Revoke information on buffers is a tri-state value:
+ *
+ * RevokeValid clear:	no cached revoke status, need to look it up
+ * RevokeValid set, Revoked clear:
+ *			buffer has not been revoked, and cancel_revoke
+ *			need do nothing.
+ * RevokeValid set, Revoked set:
+ *			buffer has been revoked.  
+ */
+
+#ifndef __KERNEL__
+#include "jfs_user.h"
+#else
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/smp_lock.h>
+#include <linux/init.h>
+#endif
+
+static kmem_cache_t *revoke_record_cache;
+static kmem_cache_t *revoke_table_cache;
+
+/* Each revoke record represents one single revoked block.  During
+   journal replay, this involves recording the transaction ID of the
+   last transaction to revoke this block. */
+
+struct jbd_revoke_record_s 
+{
+	struct list_head  hash;
+	tid_t		  sequence;	/* Used for recovery only */
+	unsigned long	  blocknr;
+};
+
+
+/* The revoke table is just a simple hash table of revoke records. */
+struct jbd_revoke_table_s
+{
+	/* It is conceivable that we might want a larger hash table
+	 * for recovery.  Must be a power of two. */
+	int		  hash_size; 
+	int		  hash_shift; 
+	struct list_head *hash_table;
+};
+
+
+#ifdef __KERNEL__
+static void write_one_revoke_record(journal_t *, transaction_t *,
+				    struct journal_head **, int *,
+				    struct jbd_revoke_record_s *);
+static void flush_descriptor(journal_t *, struct journal_head *, int);
+#endif
+
+/* Utility functions to maintain the revoke table */
+
+/* Borrowed from buffer.c: this is a tried and tested block hash function */
+static inline int hash(journal_t *journal, unsigned long block)
+{
+	struct jbd_revoke_table_s *table = journal->j_revoke;
+	int hash_shift = table->hash_shift;
+
+	return ((block << (hash_shift - 6)) ^
+		(block >> 13) ^
+		(block << (hash_shift - 12))) & (table->hash_size - 1);
+}
+
+int insert_revoke_hash(journal_t *journal, unsigned long blocknr, tid_t seq)
+{
+	struct list_head *hash_list;
+	struct jbd_revoke_record_s *record;
+
+repeat:
+	record = kmem_cache_alloc(revoke_record_cache, GFP_NOFS);
+	if (!record)
+		goto oom;
+
+	record->sequence = seq;
+	record->blocknr = blocknr;
+	hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
+	spin_lock(&journal->j_revoke_lock);
+	list_add(&record->hash, hash_list);
+	spin_unlock(&journal->j_revoke_lock);
+	return 0;
+
+oom:
+	if (!journal_oom_retry)
+		return -ENOMEM;
+	jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__);
+	yield();
+	goto repeat;
+}
+
+/* Find a revoke record in the journal's hash table. */
+
+static struct jbd_revoke_record_s *find_revoke_record(journal_t *journal,
+						      unsigned long blocknr)
+{
+	struct list_head *hash_list;
+	struct jbd_revoke_record_s *record;
+
+	hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)];
+
+	spin_lock(&journal->j_revoke_lock);
+	record = (struct jbd_revoke_record_s *) hash_list->next;
+	while (&(record->hash) != hash_list) {
+		if (record->blocknr == blocknr) {
+			spin_unlock(&journal->j_revoke_lock);
+			return record;
+		}
+		record = (struct jbd_revoke_record_s *) record->hash.next;
+	}
+	spin_unlock(&journal->j_revoke_lock);
+	return NULL;
+}
+
+int __init journal_init_revoke_caches(void)
+{
+	revoke_record_cache = kmem_cache_create("revoke_record",
+					   sizeof(struct jbd_revoke_record_s),
+					   0, SLAB_HWCACHE_ALIGN, NULL, NULL);
+	if (revoke_record_cache == 0)
+		return -ENOMEM;
+
+	revoke_table_cache = kmem_cache_create("revoke_table",
+					   sizeof(struct jbd_revoke_table_s),
+					   0, 0, NULL, NULL);
+	if (revoke_table_cache == 0) {
+		kmem_cache_destroy(revoke_record_cache);
+		revoke_record_cache = NULL;
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+void journal_destroy_revoke_caches(void)
+{
+	kmem_cache_destroy(revoke_record_cache);
+	revoke_record_cache = NULL;
+	kmem_cache_destroy(revoke_table_cache);
+	revoke_table_cache = NULL;
+}
+
+/* Initialise the revoke table for a given journal to a given size. */
+
+int journal_init_revoke(journal_t *journal, int hash_size)
+{
+	int shift, tmp;
+
+	J_ASSERT (journal->j_revoke_table[0] == NULL);
+
+	shift = 0;
+	tmp = hash_size;
+	while((tmp >>= 1UL) != 0UL)
+		shift++;
+
+	journal->j_revoke_table[0] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
+	if (!journal->j_revoke_table[0])
+		return -ENOMEM;
+	journal->j_revoke = journal->j_revoke_table[0];
+
+	/* Check that the hash_size is a power of two */
+	J_ASSERT ((hash_size & (hash_size-1)) == 0);
+
+	journal->j_revoke->hash_size = hash_size;
+
+	journal->j_revoke->hash_shift = shift;
+
+	journal->j_revoke->hash_table =
+		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
+	if (!journal->j_revoke->hash_table) {
+		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
+		journal->j_revoke = NULL;
+		return -ENOMEM;
+	}
+
+	for (tmp = 0; tmp < hash_size; tmp++)
+		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+
+	journal->j_revoke_table[1] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
+	if (!journal->j_revoke_table[1]) {
+		kfree(journal->j_revoke_table[0]->hash_table);
+		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
+		return -ENOMEM;
+	}
+
+	journal->j_revoke = journal->j_revoke_table[1];
+
+	/* Check that the hash_size is a power of two */
+	J_ASSERT ((hash_size & (hash_size-1)) == 0);
+
+	journal->j_revoke->hash_size = hash_size;
+
+	journal->j_revoke->hash_shift = shift;
+
+	journal->j_revoke->hash_table =
+		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
+	if (!journal->j_revoke->hash_table) {
+		kfree(journal->j_revoke_table[0]->hash_table);
+		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
+		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[1]);
+		journal->j_revoke = NULL;
+		return -ENOMEM;
+	}
+
+	for (tmp = 0; tmp < hash_size; tmp++)
+		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+
+	spin_lock_init(&journal->j_revoke_lock);
+
+	return 0;
+}
+
+/* Destoy a journal's revoke table.  The table must already be empty! */
+
+void journal_destroy_revoke(journal_t *journal)
+{
+	struct jbd_revoke_table_s *table;
+	struct list_head *hash_list;
+	int i;
+
+	table = journal->j_revoke_table[0];
+	if (!table)
+		return;
+
+	for (i=0; i<table->hash_size; i++) {
+		hash_list = &table->hash_table[i];
+		J_ASSERT (list_empty(hash_list));
+	}
+
+	kfree(table->hash_table);
+	kmem_cache_free(revoke_table_cache, table);
+	journal->j_revoke = NULL;
+
+	table = journal->j_revoke_table[1];
+	if (!table)
+		return;
+
+	for (i=0; i<table->hash_size; i++) {
+		hash_list = &table->hash_table[i];
+		J_ASSERT (list_empty(hash_list));
+	}
+
+	kfree(table->hash_table);
+	kmem_cache_free(revoke_table_cache, table);
+	journal->j_revoke = NULL;
+}
+
+
+#ifdef __KERNEL__
+
+/* 
+ * journal_revoke: revoke a given buffer_head from the journal.  This
+ * prevents the block from being replayed during recovery if we take a
+ * crash after this current transaction commits.  Any subsequent
+ * metadata writes of the buffer in this transaction cancel the
+ * revoke.  
+ *
+ * Note that this call may block --- it is up to the caller to make
+ * sure that there are no further calls to journal_write_metadata
+ * before the revoke is complete.  In ext3, this implies calling the
+ * revoke before clearing the block bitmap when we are deleting
+ * metadata. 
+ *
+ * Revoke performs a journal_forget on any buffer_head passed in as a
+ * parameter, but does _not_ forget the buffer_head if the bh was only
+ * found implicitly. 
+ *
+ * bh_in may not be a journalled buffer - it may have come off
+ * the hash tables without an attached journal_head.
+ *
+ * If bh_in is non-zero, journal_revoke() will decrement its b_count
+ * by one.
+ */
+
+int journal_revoke(handle_t *handle, unsigned long blocknr, 
+		   struct buffer_head *bh_in)
+{
+	struct buffer_head *bh = NULL;
+	journal_t *journal;
+	struct block_device *bdev;
+	int err;
+
+	might_sleep();
+	if (bh_in)
+		BUFFER_TRACE(bh_in, "enter");
+
+	journal = handle->h_transaction->t_journal;
+	if (!journal_set_features(journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)){
+		J_ASSERT (!"Cannot set revoke feature!");
+		return -EINVAL;
+	}
+
+	bdev = journal->j_fs_dev;
+	bh = bh_in;
+
+	if (!bh) {
+		bh = __find_get_block(bdev, blocknr, journal->j_blocksize);
+		if (bh)
+			BUFFER_TRACE(bh, "found on hash");
+	}
+#ifdef JBD_EXPENSIVE_CHECKING
+	else {
+		struct buffer_head *bh2;
+
+		/* If there is a different buffer_head lying around in
+		 * memory anywhere... */
+		bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize);
+		if (bh2) {
+			/* ... and it has RevokeValid status... */
+			if (bh2 != bh && buffer_revokevalid(bh2))
+				/* ...then it better be revoked too,
+				 * since it's illegal to create a revoke
+				 * record against a buffer_head which is
+				 * not marked revoked --- that would
+				 * risk missing a subsequent revoke
+				 * cancel. */
+				J_ASSERT_BH(bh2, buffer_revoked(bh2));
+			put_bh(bh2);
+		}
+	}
+#endif
+
+	/* We really ought not ever to revoke twice in a row without
+           first having the revoke cancelled: it's illegal to free a
+           block twice without allocating it in between! */
+	if (bh) {
+		if (!J_EXPECT_BH(bh, !buffer_revoked(bh),
+				 "inconsistent data on disk")) {
+			if (!bh_in)
+				brelse(bh);
+			return -EIO;
+		}
+		set_buffer_revoked(bh);
+		set_buffer_revokevalid(bh);
+		if (bh_in) {
+			BUFFER_TRACE(bh_in, "call journal_forget");
+			journal_forget(handle, bh_in);
+		} else {
+			BUFFER_TRACE(bh, "call brelse");
+			__brelse(bh);
+		}
+	}
+
+	jbd_debug(2, "insert revoke for block %lu, bh_in=%p\n", blocknr, bh_in);
+	err = insert_revoke_hash(journal, blocknr,
+				handle->h_transaction->t_tid);
+	BUFFER_TRACE(bh_in, "exit");
+	return err;
+}
+
+/*
+ * Cancel an outstanding revoke.  For use only internally by the
+ * journaling code (called from journal_get_write_access).
+ *
+ * We trust buffer_revoked() on the buffer if the buffer is already
+ * being journaled: if there is no revoke pending on the buffer, then we
+ * don't do anything here.
+ *
+ * This would break if it were possible for a buffer to be revoked and
+ * discarded, and then reallocated within the same transaction.  In such
+ * a case we would have lost the revoked bit, but when we arrived here
+ * the second time we would still have a pending revoke to cancel.  So,
+ * do not trust the Revoked bit on buffers unless RevokeValid is also
+ * set.
+ *
+ * The caller must have the journal locked.
+ */
+int journal_cancel_revoke(handle_t *handle, struct journal_head *jh)
+{
+	struct jbd_revoke_record_s *record;
+	journal_t *journal = handle->h_transaction->t_journal;
+	int need_cancel;
+	int did_revoke = 0;	/* akpm: debug */
+	struct buffer_head *bh = jh2bh(jh);
+
+	jbd_debug(4, "journal_head %p, cancelling revoke\n", jh);
+
+	/* Is the existing Revoke bit valid?  If so, we trust it, and
+	 * only perform the full cancel if the revoke bit is set.  If
+	 * not, we can't trust the revoke bit, and we need to do the
+	 * full search for a revoke record. */
+	if (test_set_buffer_revokevalid(bh)) {
+		need_cancel = test_clear_buffer_revoked(bh);
+	} else {
+		need_cancel = 1;
+		clear_buffer_revoked(bh);
+	}
+
+	if (need_cancel) {
+		record = find_revoke_record(journal, bh->b_blocknr);
+		if (record) {
+			jbd_debug(4, "cancelled existing revoke on "
+				  "blocknr %llu\n", (unsigned long long)bh->b_blocknr);
+			spin_lock(&journal->j_revoke_lock);
+			list_del(&record->hash);
+			spin_unlock(&journal->j_revoke_lock);
+			kmem_cache_free(revoke_record_cache, record);
+			did_revoke = 1;
+		}
+	}
+
+#ifdef JBD_EXPENSIVE_CHECKING
+	/* There better not be one left behind by now! */
+	record = find_revoke_record(journal, bh->b_blocknr);
+	J_ASSERT_JH(jh, record == NULL);
+#endif
+
+	/* Finally, have we just cleared revoke on an unhashed
+	 * buffer_head?  If so, we'd better make sure we clear the
+	 * revoked status on any hashed alias too, otherwise the revoke
+	 * state machine will get very upset later on. */
+	if (need_cancel) {
+		struct buffer_head *bh2;
+		bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size);
+		if (bh2) {
+			if (bh2 != bh)
+				clear_buffer_revoked(bh2);
+			__brelse(bh2);
+		}
+	}
+	return did_revoke;
+}
+
+/* journal_switch_revoke table select j_revoke for next transaction
+ * we do not want to suspend any processing until all revokes are
+ * written -bzzz
+ */
+void journal_switch_revoke_table(journal_t *journal)
+{
+	int i;
+
+	if (journal->j_revoke == journal->j_revoke_table[0])
+		journal->j_revoke = journal->j_revoke_table[1];
+	else
+		journal->j_revoke = journal->j_revoke_table[0];
+
+	for (i = 0; i < journal->j_revoke->hash_size; i++) 
+		INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]);
+}
+
+/*
+ * Write revoke records to the journal for all entries in the current
+ * revoke hash, deleting the entries as we go.
+ *
+ * Called with the journal lock held.
+ */
+
+void journal_write_revoke_records(journal_t *journal, 
+				  transaction_t *transaction)
+{
+	struct journal_head *descriptor;
+	struct jbd_revoke_record_s *record;
+	struct jbd_revoke_table_s *revoke;
+	struct list_head *hash_list;
+	int i, offset, count;
+
+	descriptor = NULL; 
+	offset = 0;
+	count = 0;
+
+	/* select revoke table for committing transaction */
+	revoke = journal->j_revoke == journal->j_revoke_table[0] ?
+		journal->j_revoke_table[1] : journal->j_revoke_table[0];
+
+	for (i = 0; i < revoke->hash_size; i++) {
+		hash_list = &revoke->hash_table[i];
+
+		while (!list_empty(hash_list)) {
+			record = (struct jbd_revoke_record_s *) 
+				hash_list->next;
+			write_one_revoke_record(journal, transaction,
+						&descriptor, &offset, 
+						record);
+			count++;
+			list_del(&record->hash);
+			kmem_cache_free(revoke_record_cache, record);
+		}
+	}
+	if (descriptor)
+		flush_descriptor(journal, descriptor, offset);
+	jbd_debug(1, "Wrote %d revoke records\n", count);
+}
+
+/* 
+ * Write out one revoke record.  We need to create a new descriptor
+ * block if the old one is full or if we have not already created one.  
+ */
+
+static void write_one_revoke_record(journal_t *journal, 
+				    transaction_t *transaction,
+				    struct journal_head **descriptorp, 
+				    int *offsetp,
+				    struct jbd_revoke_record_s *record)
+{
+	struct journal_head *descriptor;
+	int offset;
+	journal_header_t *header;
+
+	/* If we are already aborting, this all becomes a noop.  We
+           still need to go round the loop in
+           journal_write_revoke_records in order to free all of the
+           revoke records: only the IO to the journal is omitted. */
+	if (is_journal_aborted(journal))
+		return;
+
+	descriptor = *descriptorp;
+	offset = *offsetp;
+
+	/* Make sure we have a descriptor with space left for the record */
+	if (descriptor) {
+		if (offset == journal->j_blocksize) {
+			flush_descriptor(journal, descriptor, offset);
+			descriptor = NULL;
+		}
+	}
+
+	if (!descriptor) {
+		descriptor = journal_get_descriptor_buffer(journal);
+		if (!descriptor)
+			return;
+		header = (journal_header_t *) &jh2bh(descriptor)->b_data[0];
+		header->h_magic     = cpu_to_be32(JFS_MAGIC_NUMBER);
+		header->h_blocktype = cpu_to_be32(JFS_REVOKE_BLOCK);
+		header->h_sequence  = cpu_to_be32(transaction->t_tid);
+
+		/* Record it so that we can wait for IO completion later */
+		JBUFFER_TRACE(descriptor, "file as BJ_LogCtl");
+		journal_file_buffer(descriptor, transaction, BJ_LogCtl);
+
+		offset = sizeof(journal_revoke_header_t);
+		*descriptorp = descriptor;
+	}
+
+	* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = 
+		cpu_to_be32(record->blocknr);
+	offset += 4;
+	*offsetp = offset;
+}
+
+/* 
+ * Flush a revoke descriptor out to the journal.  If we are aborting,
+ * this is a noop; otherwise we are generating a buffer which needs to
+ * be waited for during commit, so it has to go onto the appropriate
+ * journal buffer list.
+ */
+
+static void flush_descriptor(journal_t *journal, 
+			     struct journal_head *descriptor, 
+			     int offset)
+{
+	journal_revoke_header_t *header;
+	struct buffer_head *bh = jh2bh(descriptor);
+
+	if (is_journal_aborted(journal)) {
+		put_bh(bh);
+		return;
+	}
+
+	header = (journal_revoke_header_t *) jh2bh(descriptor)->b_data;
+	header->r_count = cpu_to_be32(offset);
+	set_buffer_jwrite(bh);
+	BUFFER_TRACE(bh, "write");
+	set_buffer_dirty(bh);
+	ll_rw_block(WRITE, 1, &bh);
+}
+#endif
+
+/* 
+ * Revoke support for recovery.
+ *
+ * Recovery needs to be able to:
+ *
+ *  record all revoke records, including the tid of the latest instance
+ *  of each revoke in the journal
+ *
+ *  check whether a given block in a given transaction should be replayed
+ *  (ie. has not been revoked by a revoke record in that or a subsequent
+ *  transaction)
+ * 
+ *  empty the revoke table after recovery.
+ */
+
+/*
+ * First, setting revoke records.  We create a new revoke record for
+ * every block ever revoked in the log as we scan it for recovery, and
+ * we update the existing records if we find multiple revokes for a
+ * single block. 
+ */
+
+int journal_set_revoke(journal_t *journal, 
+		       unsigned long blocknr, 
+		       tid_t sequence)
+{
+	struct jbd_revoke_record_s *record;
+
+	record = find_revoke_record(journal, blocknr);
+	if (record) {
+		/* If we have multiple occurrences, only record the
+		 * latest sequence number in the hashed record */
+		if (tid_gt(sequence, record->sequence))
+			record->sequence = sequence;
+		return 0;
+	} 
+	return insert_revoke_hash(journal, blocknr, sequence);
+}
+
+/* 
+ * Test revoke records.  For a given block referenced in the log, has
+ * that block been revoked?  A revoke record with a given transaction
+ * sequence number revokes all blocks in that transaction and earlier
+ * ones, but later transactions still need replayed.
+ */
+
+int journal_test_revoke(journal_t *journal, 
+			unsigned long blocknr,
+			tid_t sequence)
+{
+	struct jbd_revoke_record_s *record;
+
+	record = find_revoke_record(journal, blocknr);
+	if (!record)
+		return 0;
+	if (tid_gt(sequence, record->sequence))
+		return 0;
+	return 1;
+}
+
+/*
+ * Finally, once recovery is over, we need to clear the revoke table so
+ * that it can be reused by the running filesystem.
+ */
+
+void journal_clear_revoke(journal_t *journal)
+{
+	int i;
+	struct list_head *hash_list;
+	struct jbd_revoke_record_s *record;
+	struct jbd_revoke_table_s *revoke;
+
+	revoke = journal->j_revoke;
+
+	for (i = 0; i < revoke->hash_size; i++) {
+		hash_list = &revoke->hash_table[i];
+		while (!list_empty(hash_list)) {
+			record = (struct jbd_revoke_record_s*) hash_list->next;
+			list_del(&record->hash);
+			kmem_cache_free(revoke_record_cache, record);
+		}
+	}
+}
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
new file mode 100644
index 00000000000..932e7c1ef4a
--- /dev/null
+++ b/fs/jbd/transaction.c
@@ -0,0 +1,2062 @@
+/*
+ * linux/fs/transaction.c
+ * 
+ * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
+ *
+ * Copyright 1998 Red Hat corp --- All Rights Reserved
+ *
+ * This file is part of the Linux kernel and is made available under
+ * the terms of the GNU General Public License, version 2, or at your
+ * option, any later version, incorporated herein by reference.
+ *
+ * Generic filesystem transaction handling code; part of the ext2fs
+ * journaling system.  
+ *
+ * This file manages transactions (compound commits managed by the
+ * journaling code) and handles (individual atomic operations by the
+ * filesystem).
+ */
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/jbd.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/timer.h>
+#include <linux/smp_lock.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+
+/*
+ * get_transaction: obtain a new transaction_t object.
+ *
+ * Simply allocate and initialise a new transaction.  Create it in
+ * RUNNING state and add it to the current journal (which should not
+ * have an existing running transaction: we only make a new transaction
+ * once we have started to commit the old one).
+ *
+ * Preconditions:
+ *	The journal MUST be locked.  We don't perform atomic mallocs on the
+ *	new transaction	and we can't block without protecting against other
+ *	processes trying to touch the journal while it is in transition.
+ *
+ * Called under j_state_lock
+ */
+
+static transaction_t *
+get_transaction(journal_t *journal, transaction_t *transaction)
+{
+	transaction->t_journal = journal;
+	transaction->t_state = T_RUNNING;
+	transaction->t_tid = journal->j_transaction_sequence++;
+	transaction->t_expires = jiffies + journal->j_commit_interval;
+	spin_lock_init(&transaction->t_handle_lock);
+
+	/* Set up the commit timer for the new transaction. */
+	journal->j_commit_timer->expires = transaction->t_expires;
+	add_timer(journal->j_commit_timer);
+
+	J_ASSERT(journal->j_running_transaction == NULL);
+	journal->j_running_transaction = transaction;
+
+	return transaction;
+}
+
+/*
+ * Handle management.
+ *
+ * A handle_t is an object which represents a single atomic update to a
+ * filesystem, and which tracks all of the modifications which form part
+ * of that one update.
+ */
+
+/*
+ * start_this_handle: Given a handle, deal with any locking or stalling
+ * needed to make sure that there is enough journal space for the handle
+ * to begin.  Attach the handle to a transaction and set up the
+ * transaction's buffer credits.  
+ */
+
+static int start_this_handle(journal_t *journal, handle_t *handle)
+{
+	transaction_t *transaction;
+	int needed;
+	int nblocks = handle->h_buffer_credits;
+	transaction_t *new_transaction = NULL;
+	int ret = 0;
+
+	if (nblocks > journal->j_max_transaction_buffers) {
+		printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n",
+		       current->comm, nblocks,
+		       journal->j_max_transaction_buffers);
+		ret = -ENOSPC;
+		goto out;
+	}
+
+alloc_transaction:
+	if (!journal->j_running_transaction) {
+		new_transaction = jbd_kmalloc(sizeof(*new_transaction),
+						GFP_NOFS);
+		if (!new_transaction) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		memset(new_transaction, 0, sizeof(*new_transaction));
+	}
+
+	jbd_debug(3, "New handle %p going live.\n", handle);
+
+repeat:
+
+	/*
+	 * We need to hold j_state_lock until t_updates has been incremented,
+	 * for proper journal barrier handling
+	 */
+	spin_lock(&journal->j_state_lock);
+repeat_locked:
+	if (is_journal_aborted(journal) ||
+	    (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) {
+		spin_unlock(&journal->j_state_lock);
+		ret = -EROFS; 
+		goto out;
+	}
+
+	/* Wait on the journal's transaction barrier if necessary */
+	if (journal->j_barrier_count) {
+		spin_unlock(&journal->j_state_lock);
+		wait_event(journal->j_wait_transaction_locked,
+				journal->j_barrier_count == 0);
+		goto repeat;
+	}
+
+	if (!journal->j_running_transaction) {
+		if (!new_transaction) {
+			spin_unlock(&journal->j_state_lock);
+			goto alloc_transaction;
+		}
+		get_transaction(journal, new_transaction);
+		new_transaction = NULL;
+	}
+
+	transaction = journal->j_running_transaction;
+
+	/*
+	 * If the current transaction is locked down for commit, wait for the
+	 * lock to be released.
+	 */
+	if (transaction->t_state == T_LOCKED) {
+		DEFINE_WAIT(wait);
+
+		prepare_to_wait(&journal->j_wait_transaction_locked,
+					&wait, TASK_UNINTERRUPTIBLE);
+		spin_unlock(&journal->j_state_lock);
+		schedule();
+		finish_wait(&journal->j_wait_transaction_locked, &wait);
+		goto repeat;
+	}
+
+	/*
+	 * If there is not enough space left in the log to write all potential
+	 * buffers requested by this operation, we need to stall pending a log
+	 * checkpoint to free some more log space.
+	 */
+	spin_lock(&transaction->t_handle_lock);
+	needed = transaction->t_outstanding_credits + nblocks;
+
+	if (needed > journal->j_max_transaction_buffers) {
+		/*
+		 * If the current transaction is already too large, then start
+		 * to commit it: we can then go back and attach this handle to
+		 * a new transaction.
+		 */
+		DEFINE_WAIT(wait);
+
+		jbd_debug(2, "Handle %p starting new commit...\n", handle);
+		spin_unlock(&transaction->t_handle_lock);
+		prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
+				TASK_UNINTERRUPTIBLE);
+		__log_start_commit(journal, transaction->t_tid);
+		spin_unlock(&journal->j_state_lock);
+		schedule();
+		finish_wait(&journal->j_wait_transaction_locked, &wait);
+		goto repeat;
+	}
+
+	/* 
+	 * The commit code assumes that it can get enough log space
+	 * without forcing a checkpoint.  This is *critical* for
+	 * correctness: a checkpoint of a buffer which is also
+	 * associated with a committing transaction creates a deadlock,
+	 * so commit simply cannot force through checkpoints.
+	 *
+	 * We must therefore ensure the necessary space in the journal
+	 * *before* starting to dirty potentially checkpointed buffers
+	 * in the new transaction. 
+	 *
+	 * The worst part is, any transaction currently committing can
+	 * reduce the free space arbitrarily.  Be careful to account for
+	 * those buffers when checkpointing.
+	 */
+
+	/*
+	 * @@@ AKPM: This seems rather over-defensive.  We're giving commit
+	 * a _lot_ of headroom: 1/4 of the journal plus the size of
+	 * the committing transaction.  Really, we only need to give it
+	 * committing_transaction->t_outstanding_credits plus "enough" for
+	 * the log control blocks.
+	 * Also, this test is inconsitent with the matching one in
+	 * journal_extend().
+	 */
+	if (__log_space_left(journal) < jbd_space_needed(journal)) {
+		jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle);
+		spin_unlock(&transaction->t_handle_lock);
+		__log_wait_for_space(journal);
+		goto repeat_locked;
+	}
+
+	/* OK, account for the buffers that this operation expects to
+	 * use and add the handle to the running transaction. */
+
+	handle->h_transaction = transaction;
+	transaction->t_outstanding_credits += nblocks;
+	transaction->t_updates++;
+	transaction->t_handle_count++;
+	jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n",
+		  handle, nblocks, transaction->t_outstanding_credits,
+		  __log_space_left(journal));
+	spin_unlock(&transaction->t_handle_lock);
+	spin_unlock(&journal->j_state_lock);
+out:
+	if (new_transaction)
+		kfree(new_transaction);
+	return ret;
+}
+
+/* Allocate a new handle.  This should probably be in a slab... */
+static handle_t *new_handle(int nblocks)
+{
+	handle_t *handle = jbd_alloc_handle(GFP_NOFS);
+	if (!handle)
+		return NULL;
+	memset(handle, 0, sizeof(*handle));
+	handle->h_buffer_credits = nblocks;
+	handle->h_ref = 1;
+
+	return handle;
+}
+
+/**
+ * handle_t *journal_start() - Obtain a new handle.  
+ * @journal: Journal to start transaction on.
+ * @nblocks: number of block buffer we might modify
+ *
+ * We make sure that the transaction can guarantee at least nblocks of
+ * modified buffers in the log.  We block until the log can guarantee
+ * that much space.  
+ *
+ * This function is visible to journal users (like ext3fs), so is not
+ * called with the journal already locked.
+ *
+ * Return a pointer to a newly allocated handle, or NULL on failure
+ */
+handle_t *journal_start(journal_t *journal, int nblocks)
+{
+	handle_t *handle = journal_current_handle();
+	int err;
+
+	if (!journal)
+		return ERR_PTR(-EROFS);
+
+	if (handle) {
+		J_ASSERT(handle->h_transaction->t_journal == journal);
+		handle->h_ref++;
+		return handle;
+	}
+
+	handle = new_handle(nblocks);
+	if (!handle)
+		return ERR_PTR(-ENOMEM);
+
+	current->journal_info = handle;
+
+	err = start_this_handle(journal, handle);
+	if (err < 0) {
+		jbd_free_handle(handle);
+		current->journal_info = NULL;
+		handle = ERR_PTR(err);
+	}
+	return handle;
+}
+
+/**
+ * int journal_extend() - extend buffer credits.
+ * @handle:  handle to 'extend'
+ * @nblocks: nr blocks to try to extend by.
+ * 
+ * Some transactions, such as large extends and truncates, can be done
+ * atomically all at once or in several stages.  The operation requests
+ * a credit for a number of buffer modications in advance, but can
+ * extend its credit if it needs more.  
+ *
+ * journal_extend tries to give the running handle more buffer credits.
+ * It does not guarantee that allocation - this is a best-effort only.
+ * The calling process MUST be able to deal cleanly with a failure to
+ * extend here.
+ *
+ * Return 0 on success, non-zero on failure.
+ *
+ * return code < 0 implies an error
+ * return code > 0 implies normal transaction-full status.
+ */
+int journal_extend(handle_t *handle, int nblocks)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	int result;
+	int wanted;
+
+	result = -EIO;
+	if (is_handle_aborted(handle))
+		goto out;
+
+	result = 1;
+
+	spin_lock(&journal->j_state_lock);
+
+	/* Don't extend a locked-down transaction! */
+	if (handle->h_transaction->t_state != T_RUNNING) {
+		jbd_debug(3, "denied handle %p %d blocks: "
+			  "transaction not running\n", handle, nblocks);
+		goto error_out;
+	}
+
+	spin_lock(&transaction->t_handle_lock);
+	wanted = transaction->t_outstanding_credits + nblocks;
+
+	if (wanted > journal->j_max_transaction_buffers) {
+		jbd_debug(3, "denied handle %p %d blocks: "
+			  "transaction too large\n", handle, nblocks);
+		goto unlock;
+	}
+
+	if (wanted > __log_space_left(journal)) {
+		jbd_debug(3, "denied handle %p %d blocks: "
+			  "insufficient log space\n", handle, nblocks);
+		goto unlock;
+	}
+
+	handle->h_buffer_credits += nblocks;
+	transaction->t_outstanding_credits += nblocks;
+	result = 0;
+
+	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
+unlock:
+	spin_unlock(&transaction->t_handle_lock);
+error_out:
+	spin_unlock(&journal->j_state_lock);
+out:
+	return result;
+}
+
+
+/**
+ * int journal_restart() - restart a handle .
+ * @handle:  handle to restart
+ * @nblocks: nr credits requested
+ * 
+ * Restart a handle for a multi-transaction filesystem
+ * operation.
+ *
+ * If the journal_extend() call above fails to grant new buffer credits
+ * to a running handle, a call to journal_restart will commit the
+ * handle's transaction so far and reattach the handle to a new
+ * transaction capabable of guaranteeing the requested number of
+ * credits.
+ */
+
+int journal_restart(handle_t *handle, int nblocks)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	int ret;
+
+	/* If we've had an abort of any type, don't even think about
+	 * actually doing the restart! */
+	if (is_handle_aborted(handle))
+		return 0;
+
+	/*
+	 * First unlink the handle from its current transaction, and start the
+	 * commit on that.
+	 */
+	J_ASSERT(transaction->t_updates > 0);
+	J_ASSERT(journal_current_handle() == handle);
+
+	spin_lock(&journal->j_state_lock);
+	spin_lock(&transaction->t_handle_lock);
+	transaction->t_outstanding_credits -= handle->h_buffer_credits;
+	transaction->t_updates--;
+
+	if (!transaction->t_updates)
+		wake_up(&journal->j_wait_updates);
+	spin_unlock(&transaction->t_handle_lock);
+
+	jbd_debug(2, "restarting handle %p\n", handle);
+	__log_start_commit(journal, transaction->t_tid);
+	spin_unlock(&journal->j_state_lock);
+
+	handle->h_buffer_credits = nblocks;
+	ret = start_this_handle(journal, handle);
+	return ret;
+}
+
+
+/**
+ * void journal_lock_updates () - establish a transaction barrier.
+ * @journal:  Journal to establish a barrier on.
+ *
+ * This locks out any further updates from being started, and blocks
+ * until all existing updates have completed, returning only once the
+ * journal is in a quiescent state with no updates running.
+ *
+ * The journal lock should not be held on entry.
+ */
+void journal_lock_updates(journal_t *journal)
+{
+	DEFINE_WAIT(wait);
+
+	spin_lock(&journal->j_state_lock);
+	++journal->j_barrier_count;
+
+	/* Wait until there are no running updates */
+	while (1) {
+		transaction_t *transaction = journal->j_running_transaction;
+
+		if (!transaction)
+			break;
+
+		spin_lock(&transaction->t_handle_lock);
+		if (!transaction->t_updates) {
+			spin_unlock(&transaction->t_handle_lock);
+			break;
+		}
+		prepare_to_wait(&journal->j_wait_updates, &wait,
+				TASK_UNINTERRUPTIBLE);
+		spin_unlock(&transaction->t_handle_lock);
+		spin_unlock(&journal->j_state_lock);
+		schedule();
+		finish_wait(&journal->j_wait_updates, &wait);
+		spin_lock(&journal->j_state_lock);
+	}
+	spin_unlock(&journal->j_state_lock);
+
+	/*
+	 * We have now established a barrier against other normal updates, but
+	 * we also need to barrier against other journal_lock_updates() calls
+	 * to make sure that we serialise special journal-locked operations
+	 * too.
+	 */
+	down(&journal->j_barrier);
+}
+
+/**
+ * void journal_unlock_updates (journal_t* journal) - release barrier
+ * @journal:  Journal to release the barrier on.
+ * 
+ * Release a transaction barrier obtained with journal_lock_updates().
+ *
+ * Should be called without the journal lock held.
+ */
+void journal_unlock_updates (journal_t *journal)
+{
+	J_ASSERT(journal->j_barrier_count != 0);
+
+	up(&journal->j_barrier);
+	spin_lock(&journal->j_state_lock);
+	--journal->j_barrier_count;
+	spin_unlock(&journal->j_state_lock);
+	wake_up(&journal->j_wait_transaction_locked);
+}
+
+/*
+ * Report any unexpected dirty buffers which turn up.  Normally those
+ * indicate an error, but they can occur if the user is running (say)
+ * tune2fs to modify the live filesystem, so we need the option of
+ * continuing as gracefully as possible.  #
+ *
+ * The caller should already hold the journal lock and
+ * j_list_lock spinlock: most callers will need those anyway
+ * in order to probe the buffer's journaling state safely.
+ */
+static void jbd_unexpected_dirty_buffer(struct journal_head *jh)
+{
+	struct buffer_head *bh = jh2bh(jh);
+	int jlist;
+
+	if (buffer_dirty(bh)) {
+		/* If this buffer is one which might reasonably be dirty
+		 * --- ie. data, or not part of this journal --- then
+		 * we're OK to leave it alone, but otherwise we need to
+		 * move the dirty bit to the journal's own internal
+		 * JBDDirty bit. */
+		jlist = jh->b_jlist;
+
+		if (jlist == BJ_Metadata || jlist == BJ_Reserved || 
+		    jlist == BJ_Shadow || jlist == BJ_Forget) {
+			if (test_clear_buffer_dirty(jh2bh(jh))) {
+				set_bit(BH_JBDDirty, &jh2bh(jh)->b_state);
+			}
+		}
+	}
+}
+
+/*
+ * If the buffer is already part of the current transaction, then there
+ * is nothing we need to do.  If it is already part of a prior
+ * transaction which we are still committing to disk, then we need to
+ * make sure that we do not overwrite the old copy: we do copy-out to
+ * preserve the copy going to disk.  We also account the buffer against
+ * the handle's metadata buffer credits (unless the buffer is already
+ * part of the transaction, that is).
+ *
+ */
+static int
+do_get_write_access(handle_t *handle, struct journal_head *jh,
+			int force_copy)
+{
+	struct buffer_head *bh;
+	transaction_t *transaction;
+	journal_t *journal;
+	int error;
+	char *frozen_buffer = NULL;
+	int need_copy = 0;
+
+	if (is_handle_aborted(handle))
+		return -EROFS;
+
+	transaction = handle->h_transaction;
+	journal = transaction->t_journal;
+
+	jbd_debug(5, "buffer_head %p, force_copy %d\n", jh, force_copy);
+
+	JBUFFER_TRACE(jh, "entry");
+repeat:
+	bh = jh2bh(jh);
+
+	/* @@@ Need to check for errors here at some point. */
+
+	lock_buffer(bh);
+	jbd_lock_bh_state(bh);
+
+	/* We now hold the buffer lock so it is safe to query the buffer
+	 * state.  Is the buffer dirty? 
+	 * 
+	 * If so, there are two possibilities.  The buffer may be
+	 * non-journaled, and undergoing a quite legitimate writeback.
+	 * Otherwise, it is journaled, and we don't expect dirty buffers
+	 * in that state (the buffers should be marked JBD_Dirty
+	 * instead.)  So either the IO is being done under our own
+	 * control and this is a bug, or it's a third party IO such as
+	 * dump(8) (which may leave the buffer scheduled for read ---
+	 * ie. locked but not dirty) or tune2fs (which may actually have
+	 * the buffer dirtied, ugh.)  */
+
+	if (buffer_dirty(bh)) {
+		/*
+		 * First question: is this buffer already part of the current
+		 * transaction or the existing committing transaction?
+		 */
+		if (jh->b_transaction) {
+			J_ASSERT_JH(jh,
+				jh->b_transaction == transaction || 
+				jh->b_transaction ==
+					journal->j_committing_transaction);
+			if (jh->b_next_transaction)
+				J_ASSERT_JH(jh, jh->b_next_transaction ==
+							transaction);
+			JBUFFER_TRACE(jh, "Unexpected dirty buffer");
+			jbd_unexpected_dirty_buffer(jh);
+ 		}
+ 	}
+
+	unlock_buffer(bh);
+
+	error = -EROFS;
+	if (is_handle_aborted(handle)) {
+		jbd_unlock_bh_state(bh);
+		goto out;
+	}
+	error = 0;
+
+	/*
+	 * The buffer is already part of this transaction if b_transaction or
+	 * b_next_transaction points to it
+	 */
+	if (jh->b_transaction == transaction ||
+	    jh->b_next_transaction == transaction)
+		goto done;
+
+	/*
+	 * If there is already a copy-out version of this buffer, then we don't
+	 * need to make another one
+	 */
+	if (jh->b_frozen_data) {
+		JBUFFER_TRACE(jh, "has frozen data");
+		J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+		jh->b_next_transaction = transaction;
+		goto done;
+	}
+
+	/* Is there data here we need to preserve? */
+
+	if (jh->b_transaction && jh->b_transaction != transaction) {
+		JBUFFER_TRACE(jh, "owned by older transaction");
+		J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+		J_ASSERT_JH(jh, jh->b_transaction ==
+					journal->j_committing_transaction);
+
+		/* There is one case we have to be very careful about.
+		 * If the committing transaction is currently writing
+		 * this buffer out to disk and has NOT made a copy-out,
+		 * then we cannot modify the buffer contents at all
+		 * right now.  The essence of copy-out is that it is the
+		 * extra copy, not the primary copy, which gets
+		 * journaled.  If the primary copy is already going to
+		 * disk then we cannot do copy-out here. */
+
+		if (jh->b_jlist == BJ_Shadow) {
+			DEFINE_WAIT_BIT(wait, &bh->b_state, BH_Unshadow);
+			wait_queue_head_t *wqh;
+
+			wqh = bit_waitqueue(&bh->b_state, BH_Unshadow);
+
+			JBUFFER_TRACE(jh, "on shadow: sleep");
+			jbd_unlock_bh_state(bh);
+			/* commit wakes up all shadow buffers after IO */
+			for ( ; ; ) {
+				prepare_to_wait(wqh, &wait.wait,
+						TASK_UNINTERRUPTIBLE);
+				if (jh->b_jlist != BJ_Shadow)
+					break;
+				schedule();
+			}
+			finish_wait(wqh, &wait.wait);
+			goto repeat;
+		}
+
+		/* Only do the copy if the currently-owning transaction
+		 * still needs it.  If it is on the Forget list, the
+		 * committing transaction is past that stage.  The
+		 * buffer had better remain locked during the kmalloc,
+		 * but that should be true --- we hold the journal lock
+		 * still and the buffer is already on the BUF_JOURNAL
+		 * list so won't be flushed. 
+		 *
+		 * Subtle point, though: if this is a get_undo_access,
+		 * then we will be relying on the frozen_data to contain
+		 * the new value of the committed_data record after the
+		 * transaction, so we HAVE to force the frozen_data copy
+		 * in that case. */
+
+		if (jh->b_jlist != BJ_Forget || force_copy) {
+			JBUFFER_TRACE(jh, "generate frozen data");
+			if (!frozen_buffer) {
+				JBUFFER_TRACE(jh, "allocate memory for buffer");
+				jbd_unlock_bh_state(bh);
+				frozen_buffer = jbd_kmalloc(jh2bh(jh)->b_size,
+							    GFP_NOFS);
+				if (!frozen_buffer) {
+					printk(KERN_EMERG
+					       "%s: OOM for frozen_buffer\n",
+					       __FUNCTION__);
+					JBUFFER_TRACE(jh, "oom!");
+					error = -ENOMEM;
+					jbd_lock_bh_state(bh);
+					goto done;
+				}
+				goto repeat;
+			}
+			jh->b_frozen_data = frozen_buffer;
+			frozen_buffer = NULL;
+			need_copy = 1;
+		}
+		jh->b_next_transaction = transaction;
+	}
+
+
+	/*
+	 * Finally, if the buffer is not journaled right now, we need to make
+	 * sure it doesn't get written to disk before the caller actually
+	 * commits the new data
+	 */
+	if (!jh->b_transaction) {
+		JBUFFER_TRACE(jh, "no transaction");
+		J_ASSERT_JH(jh, !jh->b_next_transaction);
+		jh->b_transaction = transaction;
+		JBUFFER_TRACE(jh, "file as BJ_Reserved");
+		spin_lock(&journal->j_list_lock);
+		__journal_file_buffer(jh, transaction, BJ_Reserved);
+		spin_unlock(&journal->j_list_lock);
+	}
+
+done:
+	if (need_copy) {
+		struct page *page;
+		int offset;
+		char *source;
+
+		J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)),
+			    "Possible IO failure.\n");
+		page = jh2bh(jh)->b_page;
+		offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK;
+		source = kmap_atomic(page, KM_USER0);
+		memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size);
+		kunmap_atomic(source, KM_USER0);
+	}
+	jbd_unlock_bh_state(bh);
+
+	/*
+	 * If we are about to journal a buffer, then any revoke pending on it is
+	 * no longer valid
+	 */
+	journal_cancel_revoke(handle, jh);
+
+out:
+	if (frozen_buffer)
+		kfree(frozen_buffer);
+
+	JBUFFER_TRACE(jh, "exit");
+	return error;
+}
+
+/**
+ * int journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update.
+ * @handle: transaction to add buffer modifications to
+ * @bh:     bh to be used for metadata writes
+ * @credits: variable that will receive credits for the buffer
+ *
+ * Returns an error code or 0 on success.
+ *
+ * In full data journalling mode the buffer may be of type BJ_AsyncData,
+ * because we're write()ing a buffer which is also part of a shared mapping.
+ */
+
+int journal_get_write_access(handle_t *handle, struct buffer_head *bh)
+{
+	struct journal_head *jh = journal_add_journal_head(bh);
+	int rc;
+
+	/* We do not want to get caught playing with fields which the
+	 * log thread also manipulates.  Make sure that the buffer
+	 * completes any outstanding IO before proceeding. */
+	rc = do_get_write_access(handle, jh, 0);
+	journal_put_journal_head(jh);
+	return rc;
+}
+
+
+/*
+ * When the user wants to journal a newly created buffer_head
+ * (ie. getblk() returned a new buffer and we are going to populate it
+ * manually rather than reading off disk), then we need to keep the
+ * buffer_head locked until it has been completely filled with new
+ * data.  In this case, we should be able to make the assertion that
+ * the bh is not already part of an existing transaction.  
+ * 
+ * The buffer should already be locked by the caller by this point.
+ * There is no lock ranking violation: it was a newly created,
+ * unlocked buffer beforehand. */
+
+/**
+ * int journal_get_create_access () - notify intent to use newly created bh
+ * @handle: transaction to new buffer to
+ * @bh: new buffer.
+ *
+ * Call this if you create a new bh.
+ */
+int journal_get_create_access(handle_t *handle, struct buffer_head *bh) 
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	struct journal_head *jh = journal_add_journal_head(bh);
+	int err;
+
+	jbd_debug(5, "journal_head %p\n", jh);
+	err = -EROFS;
+	if (is_handle_aborted(handle))
+		goto out;
+	err = 0;
+
+	JBUFFER_TRACE(jh, "entry");
+	/*
+	 * The buffer may already belong to this transaction due to pre-zeroing
+	 * in the filesystem's new_block code.  It may also be on the previous,
+	 * committing transaction's lists, but it HAS to be in Forget state in
+	 * that case: the transaction must have deleted the buffer for it to be
+	 * reused here.
+	 */
+	jbd_lock_bh_state(bh);
+	spin_lock(&journal->j_list_lock);
+	J_ASSERT_JH(jh, (jh->b_transaction == transaction ||
+		jh->b_transaction == NULL ||
+		(jh->b_transaction == journal->j_committing_transaction &&
+			  jh->b_jlist == BJ_Forget)));
+
+	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+	J_ASSERT_JH(jh, buffer_locked(jh2bh(jh)));
+
+	if (jh->b_transaction == NULL) {
+		jh->b_transaction = transaction;
+		JBUFFER_TRACE(jh, "file as BJ_Reserved");
+		__journal_file_buffer(jh, transaction, BJ_Reserved);
+	} else if (jh->b_transaction == journal->j_committing_transaction) {
+		JBUFFER_TRACE(jh, "set next transaction");
+		jh->b_next_transaction = transaction;
+	}
+	spin_unlock(&journal->j_list_lock);
+	jbd_unlock_bh_state(bh);
+
+	/*
+	 * akpm: I added this.  ext3_alloc_branch can pick up new indirect
+	 * blocks which contain freed but then revoked metadata.  We need
+	 * to cancel the revoke in case we end up freeing it yet again
+	 * and the reallocating as data - this would cause a second revoke,
+	 * which hits an assertion error.
+	 */
+	JBUFFER_TRACE(jh, "cancelling revoke");
+	journal_cancel_revoke(handle, jh);
+	journal_put_journal_head(jh);
+out:
+	return err;
+}
+
+/**
+ * int journal_get_undo_access() -  Notify intent to modify metadata with
+ *     non-rewindable consequences
+ * @handle: transaction
+ * @bh: buffer to undo
+ * @credits: store the number of taken credits here (if not NULL)
+ *
+ * Sometimes there is a need to distinguish between metadata which has
+ * been committed to disk and that which has not.  The ext3fs code uses
+ * this for freeing and allocating space, we have to make sure that we
+ * do not reuse freed space until the deallocation has been committed,
+ * since if we overwrote that space we would make the delete
+ * un-rewindable in case of a crash.
+ * 
+ * To deal with that, journal_get_undo_access requests write access to a
+ * buffer for parts of non-rewindable operations such as delete
+ * operations on the bitmaps.  The journaling code must keep a copy of
+ * the buffer's contents prior to the undo_access call until such time
+ * as we know that the buffer has definitely been committed to disk.
+ * 
+ * We never need to know which transaction the committed data is part
+ * of, buffers touched here are guaranteed to be dirtied later and so
+ * will be committed to a new transaction in due course, at which point
+ * we can discard the old committed data pointer.
+ *
+ * Returns error number or 0 on success.
+ */
+int journal_get_undo_access(handle_t *handle, struct buffer_head *bh)
+{
+	int err;
+	struct journal_head *jh = journal_add_journal_head(bh);
+	char *committed_data = NULL;
+
+	JBUFFER_TRACE(jh, "entry");
+
+	/*
+	 * Do this first --- it can drop the journal lock, so we want to
+	 * make sure that obtaining the committed_data is done
+	 * atomically wrt. completion of any outstanding commits.
+	 */
+	err = do_get_write_access(handle, jh, 1);
+	if (err)
+		goto out;
+
+repeat:
+	if (!jh->b_committed_data) {
+		committed_data = jbd_kmalloc(jh2bh(jh)->b_size, GFP_NOFS);
+		if (!committed_data) {
+			printk(KERN_EMERG "%s: No memory for committed data\n",
+				__FUNCTION__);
+			err = -ENOMEM;
+			goto out;
+		}
+	}
+
+	jbd_lock_bh_state(bh);
+	if (!jh->b_committed_data) {
+		/* Copy out the current buffer contents into the
+		 * preserved, committed copy. */
+		JBUFFER_TRACE(jh, "generate b_committed data");
+		if (!committed_data) {
+			jbd_unlock_bh_state(bh);
+			goto repeat;
+		}
+
+		jh->b_committed_data = committed_data;
+		committed_data = NULL;
+		memcpy(jh->b_committed_data, bh->b_data, bh->b_size);
+	}
+	jbd_unlock_bh_state(bh);
+out:
+	journal_put_journal_head(jh);
+	if (committed_data)
+		kfree(committed_data);
+	return err;
+}
+
+/** 
+ * int journal_dirty_data() -  mark a buffer as containing dirty data which
+ *                             needs to be flushed before we can commit the
+ *                             current transaction.  
+ * @handle: transaction
+ * @bh: bufferhead to mark
+ * 
+ * The buffer is placed on the transaction's data list and is marked as
+ * belonging to the transaction.
+ *
+ * Returns error number or 0 on success.
+ *
+ * journal_dirty_data() can be called via page_launder->ext3_writepage
+ * by kswapd.
+ */
+int journal_dirty_data(handle_t *handle, struct buffer_head *bh)
+{
+	journal_t *journal = handle->h_transaction->t_journal;
+	int need_brelse = 0;
+	struct journal_head *jh;
+
+	if (is_handle_aborted(handle))
+		return 0;
+
+	jh = journal_add_journal_head(bh);
+	JBUFFER_TRACE(jh, "entry");
+
+	/*
+	 * The buffer could *already* be dirty.  Writeout can start
+	 * at any time.
+	 */
+	jbd_debug(4, "jh: %p, tid:%d\n", jh, handle->h_transaction->t_tid);
+
+	/*
+	 * What if the buffer is already part of a running transaction?
+	 * 
+	 * There are two cases:
+	 * 1) It is part of the current running transaction.  Refile it,
+	 *    just in case we have allocated it as metadata, deallocated
+	 *    it, then reallocated it as data. 
+	 * 2) It is part of the previous, still-committing transaction.
+	 *    If all we want to do is to guarantee that the buffer will be
+	 *    written to disk before this new transaction commits, then
+	 *    being sure that the *previous* transaction has this same 
+	 *    property is sufficient for us!  Just leave it on its old
+	 *    transaction.
+	 *
+	 * In case (2), the buffer must not already exist as metadata
+	 * --- that would violate write ordering (a transaction is free
+	 * to write its data at any point, even before the previous
+	 * committing transaction has committed).  The caller must
+	 * never, ever allow this to happen: there's nothing we can do
+	 * about it in this layer.
+	 */
+	jbd_lock_bh_state(bh);
+	spin_lock(&journal->j_list_lock);
+	if (jh->b_transaction) {
+		JBUFFER_TRACE(jh, "has transaction");
+		if (jh->b_transaction != handle->h_transaction) {
+			JBUFFER_TRACE(jh, "belongs to older transaction");
+			J_ASSERT_JH(jh, jh->b_transaction ==
+					journal->j_committing_transaction);
+
+			/* @@@ IS THIS TRUE  ? */
+			/*
+			 * Not any more.  Scenario: someone does a write()
+			 * in data=journal mode.  The buffer's transaction has
+			 * moved into commit.  Then someone does another
+			 * write() to the file.  We do the frozen data copyout
+			 * and set b_next_transaction to point to j_running_t.
+			 * And while we're in that state, someone does a
+			 * writepage() in an attempt to pageout the same area
+			 * of the file via a shared mapping.  At present that
+			 * calls journal_dirty_data(), and we get right here.
+			 * It may be too late to journal the data.  Simply
+			 * falling through to the next test will suffice: the
+			 * data will be dirty and wil be checkpointed.  The
+			 * ordering comments in the next comment block still
+			 * apply.
+			 */
+			//J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
+
+			/*
+			 * If we're journalling data, and this buffer was
+			 * subject to a write(), it could be metadata, forget
+			 * or shadow against the committing transaction.  Now,
+			 * someone has dirtied the same darn page via a mapping
+			 * and it is being writepage()'d.
+			 * We *could* just steal the page from commit, with some
+			 * fancy locking there.  Instead, we just skip it -
+			 * don't tie the page's buffers to the new transaction
+			 * at all.
+			 * Implication: if we crash before the writepage() data
+			 * is written into the filesystem, recovery will replay
+			 * the write() data.
+			 */
+			if (jh->b_jlist != BJ_None &&
+					jh->b_jlist != BJ_SyncData &&
+					jh->b_jlist != BJ_Locked) {
+				JBUFFER_TRACE(jh, "Not stealing");
+				goto no_journal;
+			}
+
+			/*
+			 * This buffer may be undergoing writeout in commit.  We
+			 * can't return from here and let the caller dirty it
+			 * again because that can cause the write-out loop in
+			 * commit to never terminate.
+			 */
+			if (buffer_dirty(bh)) {
+				get_bh(bh);
+				spin_unlock(&journal->j_list_lock);
+				jbd_unlock_bh_state(bh);
+				need_brelse = 1;
+				sync_dirty_buffer(bh);
+				jbd_lock_bh_state(bh);
+				spin_lock(&journal->j_list_lock);
+				/* The buffer may become locked again at any
+				   time if it is redirtied */
+			}
+
+			/* journal_clean_data_list() may have got there first */
+			if (jh->b_transaction != NULL) {
+				JBUFFER_TRACE(jh, "unfile from commit");
+				__journal_temp_unlink_buffer(jh);
+				/* It still points to the committing
+				 * transaction; move it to this one so
+				 * that the refile assert checks are
+				 * happy. */
+				jh->b_transaction = handle->h_transaction;
+			}
+			/* The buffer will be refiled below */
+
+		}
+		/*
+		 * Special case --- the buffer might actually have been
+		 * allocated and then immediately deallocated in the previous,
+		 * committing transaction, so might still be left on that
+		 * transaction's metadata lists.
+		 */
+		if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) {
+			JBUFFER_TRACE(jh, "not on correct data list: unfile");
+			J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow);
+			__journal_temp_unlink_buffer(jh);
+			jh->b_transaction = handle->h_transaction;
+			JBUFFER_TRACE(jh, "file as data");
+			__journal_file_buffer(jh, handle->h_transaction,
+						BJ_SyncData);
+		}
+	} else {
+		JBUFFER_TRACE(jh, "not on a transaction");
+		__journal_file_buffer(jh, handle->h_transaction, BJ_SyncData);
+	}
+no_journal:
+	spin_unlock(&journal->j_list_lock);
+	jbd_unlock_bh_state(bh);
+	if (need_brelse) {
+		BUFFER_TRACE(bh, "brelse");
+		__brelse(bh);
+	}
+	JBUFFER_TRACE(jh, "exit");
+	journal_put_journal_head(jh);
+	return 0;
+}
+
+/** 
+ * int journal_dirty_metadata() -  mark a buffer as containing dirty metadata
+ * @handle: transaction to add buffer to.
+ * @bh: buffer to mark 
+ * 
+ * mark dirty metadata which needs to be journaled as part of the current
+ * transaction.
+ *
+ * The buffer is placed on the transaction's metadata list and is marked
+ * as belonging to the transaction.  
+ *
+ * Returns error number or 0 on success.  
+ *
+ * Special care needs to be taken if the buffer already belongs to the
+ * current committing transaction (in which case we should have frozen
+ * data present for that commit).  In that case, we don't relink the
+ * buffer: that only gets done when the old transaction finally
+ * completes its commit.
+ */
+int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	struct journal_head *jh = bh2jh(bh);
+
+	jbd_debug(5, "journal_head %p\n", jh);
+	JBUFFER_TRACE(jh, "entry");
+	if (is_handle_aborted(handle))
+		goto out;
+
+	jbd_lock_bh_state(bh);
+
+	if (jh->b_modified == 0) {
+		/*
+		 * This buffer's got modified and becoming part
+		 * of the transaction. This needs to be done
+		 * once a transaction -bzzz
+		 */
+		jh->b_modified = 1;
+		J_ASSERT_JH(jh, handle->h_buffer_credits > 0);
+		handle->h_buffer_credits--;
+	}
+
+	/*
+	 * fastpath, to avoid expensive locking.  If this buffer is already
+	 * on the running transaction's metadata list there is nothing to do.
+	 * Nobody can take it off again because there is a handle open.
+	 * I _think_ we're OK here with SMP barriers - a mistaken decision will
+	 * result in this test being false, so we go in and take the locks.
+	 */
+	if (jh->b_transaction == transaction && jh->b_jlist == BJ_Metadata) {
+		JBUFFER_TRACE(jh, "fastpath");
+		J_ASSERT_JH(jh, jh->b_transaction ==
+					journal->j_running_transaction);
+		goto out_unlock_bh;
+	}
+
+	set_buffer_jbddirty(bh);
+
+	/* 
+	 * Metadata already on the current transaction list doesn't
+	 * need to be filed.  Metadata on another transaction's list must
+	 * be committing, and will be refiled once the commit completes:
+	 * leave it alone for now. 
+	 */
+	if (jh->b_transaction != transaction) {
+		JBUFFER_TRACE(jh, "already on other transaction");
+		J_ASSERT_JH(jh, jh->b_transaction ==
+					journal->j_committing_transaction);
+		J_ASSERT_JH(jh, jh->b_next_transaction == transaction);
+		/* And this case is illegal: we can't reuse another
+		 * transaction's data buffer, ever. */
+		goto out_unlock_bh;
+	}
+
+	/* That test should have eliminated the following case: */
+	J_ASSERT_JH(jh, jh->b_frozen_data == 0);
+
+	JBUFFER_TRACE(jh, "file as BJ_Metadata");
+	spin_lock(&journal->j_list_lock);
+	__journal_file_buffer(jh, handle->h_transaction, BJ_Metadata);
+	spin_unlock(&journal->j_list_lock);
+out_unlock_bh:
+	jbd_unlock_bh_state(bh);
+out:
+	JBUFFER_TRACE(jh, "exit");
+	return 0;
+}
+
+/* 
+ * journal_release_buffer: undo a get_write_access without any buffer
+ * updates, if the update decided in the end that it didn't need access.
+ *
+ */
+void
+journal_release_buffer(handle_t *handle, struct buffer_head *bh)
+{
+	BUFFER_TRACE(bh, "entry");
+}
+
+/** 
+ * void journal_forget() - bforget() for potentially-journaled buffers.
+ * @handle: transaction handle
+ * @bh:     bh to 'forget'
+ *
+ * We can only do the bforget if there are no commits pending against the
+ * buffer.  If the buffer is dirty in the current running transaction we
+ * can safely unlink it. 
+ *
+ * bh may not be a journalled buffer at all - it may be a non-JBD
+ * buffer which came off the hashtable.  Check for this.
+ *
+ * Decrements bh->b_count by one.
+ * 
+ * Allow this call even if the handle has aborted --- it may be part of
+ * the caller's cleanup after an abort.
+ */
+int journal_forget (handle_t *handle, struct buffer_head *bh)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	struct journal_head *jh;
+	int drop_reserve = 0;
+	int err = 0;
+
+	BUFFER_TRACE(bh, "entry");
+
+	jbd_lock_bh_state(bh);
+	spin_lock(&journal->j_list_lock);
+
+	if (!buffer_jbd(bh))
+		goto not_jbd;
+	jh = bh2jh(bh);
+
+	/* Critical error: attempting to delete a bitmap buffer, maybe?
+	 * Don't do any jbd operations, and return an error. */
+	if (!J_EXPECT_JH(jh, !jh->b_committed_data,
+			 "inconsistent data on disk")) {
+		err = -EIO;
+		goto not_jbd;
+	}
+
+	/*
+	 * The buffer's going from the transaction, we must drop
+	 * all references -bzzz
+	 */
+	jh->b_modified = 0;
+
+	if (jh->b_transaction == handle->h_transaction) {
+		J_ASSERT_JH(jh, !jh->b_frozen_data);
+
+		/* If we are forgetting a buffer which is already part
+		 * of this transaction, then we can just drop it from
+		 * the transaction immediately. */
+		clear_buffer_dirty(bh);
+		clear_buffer_jbddirty(bh);
+
+		JBUFFER_TRACE(jh, "belongs to current transaction: unfile");
+
+		drop_reserve = 1;
+
+		/* 
+		 * We are no longer going to journal this buffer.
+		 * However, the commit of this transaction is still
+		 * important to the buffer: the delete that we are now
+		 * processing might obsolete an old log entry, so by
+		 * committing, we can satisfy the buffer's checkpoint.
+		 *
+		 * So, if we have a checkpoint on the buffer, we should
+		 * now refile the buffer on our BJ_Forget list so that
+		 * we know to remove the checkpoint after we commit. 
+		 */
+
+		if (jh->b_cp_transaction) {
+			__journal_temp_unlink_buffer(jh);
+			__journal_file_buffer(jh, transaction, BJ_Forget);
+		} else {
+			__journal_unfile_buffer(jh);
+			journal_remove_journal_head(bh);
+			__brelse(bh);
+			if (!buffer_jbd(bh)) {
+				spin_unlock(&journal->j_list_lock);
+				jbd_unlock_bh_state(bh);
+				__bforget(bh);
+				goto drop;
+			}
+		}
+	} else if (jh->b_transaction) {
+		J_ASSERT_JH(jh, (jh->b_transaction == 
+				 journal->j_committing_transaction));
+		/* However, if the buffer is still owned by a prior
+		 * (committing) transaction, we can't drop it yet... */
+		JBUFFER_TRACE(jh, "belongs to older transaction");
+		/* ... but we CAN drop it from the new transaction if we
+		 * have also modified it since the original commit. */
+
+		if (jh->b_next_transaction) {
+			J_ASSERT(jh->b_next_transaction == transaction);
+			jh->b_next_transaction = NULL;
+			drop_reserve = 1;
+		}
+	}
+
+not_jbd:
+	spin_unlock(&journal->j_list_lock);
+	jbd_unlock_bh_state(bh);
+	__brelse(bh);
+drop:
+	if (drop_reserve) {
+		/* no need to reserve log space for this block -bzzz */
+		handle->h_buffer_credits++;
+	}
+	return err;
+}
+
+/**
+ * int journal_stop() - complete a transaction
+ * @handle: tranaction to complete.
+ * 
+ * All done for a particular handle.
+ *
+ * There is not much action needed here.  We just return any remaining
+ * buffer credits to the transaction and remove the handle.  The only
+ * complication is that we need to start a commit operation if the
+ * filesystem is marked for synchronous update.
+ *
+ * journal_stop itself will not usually return an error, but it may
+ * do so in unusual circumstances.  In particular, expect it to 
+ * return -EIO if a journal_abort has been executed since the
+ * transaction began.
+ */
+int journal_stop(handle_t *handle)
+{
+	transaction_t *transaction = handle->h_transaction;
+	journal_t *journal = transaction->t_journal;
+	int old_handle_count, err;
+
+	J_ASSERT(transaction->t_updates > 0);
+	J_ASSERT(journal_current_handle() == handle);
+
+	if (is_handle_aborted(handle))
+		err = -EIO;
+	else
+		err = 0;
+
+	if (--handle->h_ref > 0) {
+		jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1,
+			  handle->h_ref);
+		return err;
+	}
+
+	jbd_debug(4, "Handle %p going down\n", handle);
+
+	/*
+	 * Implement synchronous transaction batching.  If the handle
+	 * was synchronous, don't force a commit immediately.  Let's
+	 * yield and let another thread piggyback onto this transaction.
+	 * Keep doing that while new threads continue to arrive.
+	 * It doesn't cost much - we're about to run a commit and sleep
+	 * on IO anyway.  Speeds up many-threaded, many-dir operations
+	 * by 30x or more...
+	 */
+	if (handle->h_sync) {
+		do {
+			old_handle_count = transaction->t_handle_count;
+			set_current_state(TASK_UNINTERRUPTIBLE);
+			schedule_timeout(1);
+		} while (old_handle_count != transaction->t_handle_count);
+	}
+
+	current->journal_info = NULL;
+	spin_lock(&journal->j_state_lock);
+	spin_lock(&transaction->t_handle_lock);
+	transaction->t_outstanding_credits -= handle->h_buffer_credits;
+	transaction->t_updates--;
+	if (!transaction->t_updates) {
+		wake_up(&journal->j_wait_updates);
+		if (journal->j_barrier_count)
+			wake_up(&journal->j_wait_transaction_locked);
+	}
+
+	/*
+	 * If the handle is marked SYNC, we need to set another commit
+	 * going!  We also want to force a commit if the current
+	 * transaction is occupying too much of the log, or if the
+	 * transaction is too old now.
+	 */
+	if (handle->h_sync ||
+			transaction->t_outstanding_credits >
+				journal->j_max_transaction_buffers ||
+	    		time_after_eq(jiffies, transaction->t_expires)) {
+		/* Do this even for aborted journals: an abort still
+		 * completes the commit thread, it just doesn't write
+		 * anything to disk. */
+		tid_t tid = transaction->t_tid;
+
+		spin_unlock(&transaction->t_handle_lock);
+		jbd_debug(2, "transaction too old, requesting commit for "
+					"handle %p\n", handle);
+		/* This is non-blocking */
+		__log_start_commit(journal, transaction->t_tid);
+		spin_unlock(&journal->j_state_lock);
+
+		/*
+		 * Special case: JFS_SYNC synchronous updates require us
+		 * to wait for the commit to complete.  
+		 */
+		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
+			err = log_wait_commit(journal, tid);
+	} else {
+		spin_unlock(&transaction->t_handle_lock);
+		spin_unlock(&journal->j_state_lock);
+	}
+
+	jbd_free_handle(handle);
+	return err;
+}
+
+/**int journal_force_commit() - force any uncommitted transactions
+ * @journal: journal to force
+ *
+ * For synchronous operations: force any uncommitted transactions
+ * to disk.  May seem kludgy, but it reuses all the handle batching
+ * code in a very simple manner.
+ */
+int journal_force_commit(journal_t *journal)
+{
+	handle_t *handle;
+	int ret;
+
+	handle = journal_start(journal, 1);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+	} else {
+		handle->h_sync = 1;
+		ret = journal_stop(handle);
+	}
+	return ret;
+}
+
+/*
+ *
+ * List management code snippets: various functions for manipulating the
+ * transaction buffer lists.
+ *
+ */
+
+/*
+ * Append a buffer to a transaction list, given the transaction's list head
+ * pointer.
+ *
+ * j_list_lock is held.
+ *
+ * jbd_lock_bh_state(jh2bh(jh)) is held.
+ */
+
+static inline void 
+__blist_add_buffer(struct journal_head **list, struct journal_head *jh)
+{
+	if (!*list) {
+		jh->b_tnext = jh->b_tprev = jh;
+		*list = jh;
+	} else {
+		/* Insert at the tail of the list to preserve order */
+		struct journal_head *first = *list, *last = first->b_tprev;
+		jh->b_tprev = last;
+		jh->b_tnext = first;
+		last->b_tnext = first->b_tprev = jh;
+	}
+}
+
+/* 
+ * Remove a buffer from a transaction list, given the transaction's list
+ * head pointer.
+ *
+ * Called with j_list_lock held, and the journal may not be locked.
+ *
+ * jbd_lock_bh_state(jh2bh(jh)) is held.
+ */
+
+static inline void
+__blist_del_buffer(struct journal_head **list, struct journal_head *jh)
+{
+	if (*list == jh) {
+		*list = jh->b_tnext;
+		if (*list == jh)
+			*list = NULL;
+	}
+	jh->b_tprev->b_tnext = jh->b_tnext;
+	jh->b_tnext->b_tprev = jh->b_tprev;
+}
+
+/* 
+ * Remove a buffer from the appropriate transaction list.
+ *
+ * Note that this function can *change* the value of
+ * bh->b_transaction->t_sync_datalist, t_buffers, t_forget,
+ * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list.  If the caller
+ * is holding onto a copy of one of thee pointers, it could go bad.
+ * Generally the caller needs to re-read the pointer from the transaction_t.
+ *
+ * Called under j_list_lock.  The journal may not be locked.
+ */
+void __journal_temp_unlink_buffer(struct journal_head *jh)
+{
+	struct journal_head **list = NULL;
+	transaction_t *transaction;
+	struct buffer_head *bh = jh2bh(jh);
+
+	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	transaction = jh->b_transaction;
+	if (transaction)
+		assert_spin_locked(&transaction->t_journal->j_list_lock);
+
+	J_ASSERT_JH(jh, jh->b_jlist < BJ_Types);
+	if (jh->b_jlist != BJ_None)
+		J_ASSERT_JH(jh, transaction != 0);
+
+	switch (jh->b_jlist) {
+	case BJ_None:
+		return;
+	case BJ_SyncData:
+		list = &transaction->t_sync_datalist;
+		break;
+	case BJ_Metadata:
+		transaction->t_nr_buffers--;
+		J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0);
+		list = &transaction->t_buffers;
+		break;
+	case BJ_Forget:
+		list = &transaction->t_forget;
+		break;
+	case BJ_IO:
+		list = &transaction->t_iobuf_list;
+		break;
+	case BJ_Shadow:
+		list = &transaction->t_shadow_list;
+		break;
+	case BJ_LogCtl:
+		list = &transaction->t_log_list;
+		break;
+	case BJ_Reserved:
+		list = &transaction->t_reserved_list;
+		break;
+	case BJ_Locked:
+		list = &transaction->t_locked_list;
+		break;
+	}
+
+	__blist_del_buffer(list, jh);
+	jh->b_jlist = BJ_None;
+	if (test_clear_buffer_jbddirty(bh))
+		mark_buffer_dirty(bh);	/* Expose it to the VM */
+}
+
+void __journal_unfile_buffer(struct journal_head *jh)
+{
+	__journal_temp_unlink_buffer(jh);
+	jh->b_transaction = NULL;
+}
+
+void journal_unfile_buffer(journal_t *journal, struct journal_head *jh)
+{
+	jbd_lock_bh_state(jh2bh(jh));
+	spin_lock(&journal->j_list_lock);
+	__journal_unfile_buffer(jh);
+	spin_unlock(&journal->j_list_lock);
+	jbd_unlock_bh_state(jh2bh(jh));
+}
+
+/*
+ * Called from journal_try_to_free_buffers().
+ *
+ * Called under jbd_lock_bh_state(bh)
+ */
+static void
+__journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
+{
+	struct journal_head *jh;
+
+	jh = bh2jh(bh);
+
+	if (buffer_locked(bh) || buffer_dirty(bh))
+		goto out;
+
+	if (jh->b_next_transaction != 0)
+		goto out;
+
+	spin_lock(&journal->j_list_lock);
+	if (jh->b_transaction != 0 && jh->b_cp_transaction == 0) {
+		if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) {
+			/* A written-back ordered data buffer */
+			JBUFFER_TRACE(jh, "release data");
+			__journal_unfile_buffer(jh);
+			journal_remove_journal_head(bh);
+			__brelse(bh);
+		}
+	} else if (jh->b_cp_transaction != 0 && jh->b_transaction == 0) {
+		/* written-back checkpointed metadata buffer */
+		if (jh->b_jlist == BJ_None) {
+			JBUFFER_TRACE(jh, "remove from checkpoint list");
+			__journal_remove_checkpoint(jh);
+			journal_remove_journal_head(bh);
+			__brelse(bh);
+		}
+	}
+	spin_unlock(&journal->j_list_lock);
+out:
+	return;
+}
+
+
+/** 
+ * int journal_try_to_free_buffers() - try to free page buffers.
+ * @journal: journal for operation
+ * @page: to try and free
+ * @unused_gfp_mask: unused
+ *
+ * 
+ * For all the buffers on this page,
+ * if they are fully written out ordered data, move them onto BUF_CLEAN
+ * so try_to_free_buffers() can reap them.
+ * 
+ * This function returns non-zero if we wish try_to_free_buffers()
+ * to be called. We do this if the page is releasable by try_to_free_buffers().
+ * We also do it if the page has locked or dirty buffers and the caller wants
+ * us to perform sync or async writeout.
+ *
+ * This complicates JBD locking somewhat.  We aren't protected by the
+ * BKL here.  We wish to remove the buffer from its committing or
+ * running transaction's ->t_datalist via __journal_unfile_buffer.
+ *
+ * This may *change* the value of transaction_t->t_datalist, so anyone
+ * who looks at t_datalist needs to lock against this function.
+ *
+ * Even worse, someone may be doing a journal_dirty_data on this
+ * buffer.  So we need to lock against that.  journal_dirty_data()
+ * will come out of the lock with the buffer dirty, which makes it
+ * ineligible for release here.
+ *
+ * Who else is affected by this?  hmm...  Really the only contender
+ * is do_get_write_access() - it could be looking at the buffer while
+ * journal_try_to_free_buffer() is changing its state.  But that
+ * cannot happen because we never reallocate freed data as metadata
+ * while the data is part of a transaction.  Yes?
+ */
+int journal_try_to_free_buffers(journal_t *journal, 
+				struct page *page, int unused_gfp_mask)
+{
+	struct buffer_head *head;
+	struct buffer_head *bh;
+	int ret = 0;
+
+	J_ASSERT(PageLocked(page));
+
+	head = page_buffers(page);
+	bh = head;
+	do {
+		struct journal_head *jh;
+
+		/*
+		 * We take our own ref against the journal_head here to avoid
+		 * having to add tons of locking around each instance of
+		 * journal_remove_journal_head() and journal_put_journal_head().
+		 */
+		jh = journal_grab_journal_head(bh);
+		if (!jh)
+			continue;
+
+		jbd_lock_bh_state(bh);
+		__journal_try_to_free_buffer(journal, bh);
+		journal_put_journal_head(jh);
+		jbd_unlock_bh_state(bh);
+		if (buffer_jbd(bh))
+			goto busy;
+	} while ((bh = bh->b_this_page) != head);
+	ret = try_to_free_buffers(page);
+busy:
+	return ret;
+}
+
+/*
+ * This buffer is no longer needed.  If it is on an older transaction's
+ * checkpoint list we need to record it on this transaction's forget list
+ * to pin this buffer (and hence its checkpointing transaction) down until
+ * this transaction commits.  If the buffer isn't on a checkpoint list, we
+ * release it.
+ * Returns non-zero if JBD no longer has an interest in the buffer.
+ *
+ * Called under j_list_lock.
+ *
+ * Called under jbd_lock_bh_state(bh).
+ */
+static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
+{
+	int may_free = 1;
+	struct buffer_head *bh = jh2bh(jh);
+
+	__journal_unfile_buffer(jh);
+
+	if (jh->b_cp_transaction) {
+		JBUFFER_TRACE(jh, "on running+cp transaction");
+		__journal_file_buffer(jh, transaction, BJ_Forget);
+		clear_buffer_jbddirty(bh);
+		may_free = 0;
+	} else {
+		JBUFFER_TRACE(jh, "on running transaction");
+		journal_remove_journal_head(bh);
+		__brelse(bh);
+	}
+	return may_free;
+}
+
+/*
+ * journal_invalidatepage 
+ *
+ * This code is tricky.  It has a number of cases to deal with.
+ *
+ * There are two invariants which this code relies on:
+ *
+ * i_size must be updated on disk before we start calling invalidatepage on the
+ * data.
+ * 
+ *  This is done in ext3 by defining an ext3_setattr method which
+ *  updates i_size before truncate gets going.  By maintaining this
+ *  invariant, we can be sure that it is safe to throw away any buffers
+ *  attached to the current transaction: once the transaction commits,
+ *  we know that the data will not be needed.
+ * 
+ *  Note however that we can *not* throw away data belonging to the
+ *  previous, committing transaction!  
+ *
+ * Any disk blocks which *are* part of the previous, committing
+ * transaction (and which therefore cannot be discarded immediately) are
+ * not going to be reused in the new running transaction
+ *
+ *  The bitmap committed_data images guarantee this: any block which is
+ *  allocated in one transaction and removed in the next will be marked
+ *  as in-use in the committed_data bitmap, so cannot be reused until
+ *  the next transaction to delete the block commits.  This means that
+ *  leaving committing buffers dirty is quite safe: the disk blocks
+ *  cannot be reallocated to a different file and so buffer aliasing is
+ *  not possible.
+ *
+ *
+ * The above applies mainly to ordered data mode.  In writeback mode we
+ * don't make guarantees about the order in which data hits disk --- in
+ * particular we don't guarantee that new dirty data is flushed before
+ * transaction commit --- so it is always safe just to discard data
+ * immediately in that mode.  --sct 
+ */
+
+/*
+ * The journal_unmap_buffer helper function returns zero if the buffer
+ * concerned remains pinned as an anonymous buffer belonging to an older
+ * transaction.
+ *
+ * We're outside-transaction here.  Either or both of j_running_transaction
+ * and j_committing_transaction may be NULL.
+ */
+static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
+{
+	transaction_t *transaction;
+	struct journal_head *jh;
+	int may_free = 1;
+	int ret;
+
+	BUFFER_TRACE(bh, "entry");
+
+	/*
+	 * It is safe to proceed here without the j_list_lock because the
+	 * buffers cannot be stolen by try_to_free_buffers as long as we are
+	 * holding the page lock. --sct
+	 */
+
+	if (!buffer_jbd(bh))
+		goto zap_buffer_unlocked;
+
+	spin_lock(&journal->j_state_lock);
+	jbd_lock_bh_state(bh);
+	spin_lock(&journal->j_list_lock);
+
+	jh = journal_grab_journal_head(bh);
+	if (!jh)
+		goto zap_buffer_no_jh;
+
+	transaction = jh->b_transaction;
+	if (transaction == NULL) {
+		/* First case: not on any transaction.  If it
+		 * has no checkpoint link, then we can zap it:
+		 * it's a writeback-mode buffer so we don't care
+		 * if it hits disk safely. */
+		if (!jh->b_cp_transaction) {
+			JBUFFER_TRACE(jh, "not on any transaction: zap");
+			goto zap_buffer;
+		}
+
+		if (!buffer_dirty(bh)) {
+			/* bdflush has written it.  We can drop it now */
+			goto zap_buffer;
+		}
+
+		/* OK, it must be in the journal but still not
+		 * written fully to disk: it's metadata or
+		 * journaled data... */
+
+		if (journal->j_running_transaction) {
+			/* ... and once the current transaction has
+			 * committed, the buffer won't be needed any
+			 * longer. */
+			JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget");
+			ret = __dispose_buffer(jh,
+					journal->j_running_transaction);
+			journal_put_journal_head(jh);
+			spin_unlock(&journal->j_list_lock);
+			jbd_unlock_bh_state(bh);
+			spin_unlock(&journal->j_state_lock);
+			return ret;
+		} else {
+			/* There is no currently-running transaction. So the
+			 * orphan record which we wrote for this file must have
+			 * passed into commit.  We must attach this buffer to
+			 * the committing transaction, if it exists. */
+			if (journal->j_committing_transaction) {
+				JBUFFER_TRACE(jh, "give to committing trans");
+				ret = __dispose_buffer(jh,
+					journal->j_committing_transaction);
+				journal_put_journal_head(jh);
+				spin_unlock(&journal->j_list_lock);
+				jbd_unlock_bh_state(bh);
+				spin_unlock(&journal->j_state_lock);
+				return ret;
+			} else {
+				/* The orphan record's transaction has
+				 * committed.  We can cleanse this buffer */
+				clear_buffer_jbddirty(bh);
+				goto zap_buffer;
+			}
+		}
+	} else if (transaction == journal->j_committing_transaction) {
+		/* If it is committing, we simply cannot touch it.  We
+		 * can remove it's next_transaction pointer from the
+		 * running transaction if that is set, but nothing
+		 * else. */
+		JBUFFER_TRACE(jh, "on committing transaction");
+		set_buffer_freed(bh);
+		if (jh->b_next_transaction) {
+			J_ASSERT(jh->b_next_transaction ==
+					journal->j_running_transaction);
+			jh->b_next_transaction = NULL;
+		}
+		journal_put_journal_head(jh);
+		spin_unlock(&journal->j_list_lock);
+		jbd_unlock_bh_state(bh);
+		spin_unlock(&journal->j_state_lock);
+		return 0;
+	} else {
+		/* Good, the buffer belongs to the running transaction.
+		 * We are writing our own transaction's data, not any
+		 * previous one's, so it is safe to throw it away
+		 * (remember that we expect the filesystem to have set
+		 * i_size already for this truncate so recovery will not
+		 * expose the disk blocks we are discarding here.) */
+		J_ASSERT_JH(jh, transaction == journal->j_running_transaction);
+		may_free = __dispose_buffer(jh, transaction);
+	}
+
+zap_buffer:
+	journal_put_journal_head(jh);
+zap_buffer_no_jh:
+	spin_unlock(&journal->j_list_lock);
+	jbd_unlock_bh_state(bh);
+	spin_unlock(&journal->j_state_lock);
+zap_buffer_unlocked:
+	clear_buffer_dirty(bh);
+	J_ASSERT_BH(bh, !buffer_jbddirty(bh));
+	clear_buffer_mapped(bh);
+	clear_buffer_req(bh);
+	clear_buffer_new(bh);
+	bh->b_bdev = NULL;
+	return may_free;
+}
+
+/** 
+ * int journal_invalidatepage() 
+ * @journal: journal to use for flush... 
+ * @page:    page to flush
+ * @offset:  length of page to invalidate.
+ *
+ * Reap page buffers containing data after offset in page.
+ *
+ * Return non-zero if the page's buffers were successfully reaped.
+ */
+int journal_invalidatepage(journal_t *journal, 
+		      struct page *page, 
+		      unsigned long offset)
+{
+	struct buffer_head *head, *bh, *next;
+	unsigned int curr_off = 0;
+	int may_free = 1;
+
+	if (!PageLocked(page))
+		BUG();
+	if (!page_has_buffers(page))
+		return 1;
+
+	/* We will potentially be playing with lists other than just the
+	 * data lists (especially for journaled data mode), so be
+	 * cautious in our locking. */
+
+	head = bh = page_buffers(page);
+	do {
+		unsigned int next_off = curr_off + bh->b_size;
+		next = bh->b_this_page;
+
+		/* AKPM: doing lock_buffer here may be overly paranoid */
+		if (offset <= curr_off) {
+		 	/* This block is wholly outside the truncation point */
+			lock_buffer(bh);
+			may_free &= journal_unmap_buffer(journal, bh);
+			unlock_buffer(bh);
+		}
+		curr_off = next_off;
+		bh = next;
+
+	} while (bh != head);
+
+	if (!offset) {
+		if (!may_free || !try_to_free_buffers(page))
+			return 0;
+		J_ASSERT(!page_has_buffers(page));
+	}
+	return 1;
+}
+
+/* 
+ * File a buffer on the given transaction list. 
+ */
+void __journal_file_buffer(struct journal_head *jh,
+			transaction_t *transaction, int jlist)
+{
+	struct journal_head **list = NULL;
+	int was_dirty = 0;
+	struct buffer_head *bh = jh2bh(jh);
+
+	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	assert_spin_locked(&transaction->t_journal->j_list_lock);
+
+	J_ASSERT_JH(jh, jh->b_jlist < BJ_Types);
+	J_ASSERT_JH(jh, jh->b_transaction == transaction ||
+				jh->b_transaction == 0);
+
+	if (jh->b_transaction && jh->b_jlist == jlist)
+		return;
+
+	/* The following list of buffer states needs to be consistent
+	 * with __jbd_unexpected_dirty_buffer()'s handling of dirty
+	 * state. */
+
+	if (jlist == BJ_Metadata || jlist == BJ_Reserved || 
+	    jlist == BJ_Shadow || jlist == BJ_Forget) {
+		if (test_clear_buffer_dirty(bh) ||
+		    test_clear_buffer_jbddirty(bh))
+			was_dirty = 1;
+	}
+
+	if (jh->b_transaction)
+		__journal_temp_unlink_buffer(jh);
+	jh->b_transaction = transaction;
+
+	switch (jlist) {
+	case BJ_None:
+		J_ASSERT_JH(jh, !jh->b_committed_data);
+		J_ASSERT_JH(jh, !jh->b_frozen_data);
+		return;
+	case BJ_SyncData:
+		list = &transaction->t_sync_datalist;
+		break;
+	case BJ_Metadata:
+		transaction->t_nr_buffers++;
+		list = &transaction->t_buffers;
+		break;
+	case BJ_Forget:
+		list = &transaction->t_forget;
+		break;
+	case BJ_IO:
+		list = &transaction->t_iobuf_list;
+		break;
+	case BJ_Shadow:
+		list = &transaction->t_shadow_list;
+		break;
+	case BJ_LogCtl:
+		list = &transaction->t_log_list;
+		break;
+	case BJ_Reserved:
+		list = &transaction->t_reserved_list;
+		break;
+	case BJ_Locked:
+		list =  &transaction->t_locked_list;
+		break;
+	}
+
+	__blist_add_buffer(list, jh);
+	jh->b_jlist = jlist;
+
+	if (was_dirty)
+		set_buffer_jbddirty(bh);
+}
+
+void journal_file_buffer(struct journal_head *jh,
+				transaction_t *transaction, int jlist)
+{
+	jbd_lock_bh_state(jh2bh(jh));
+	spin_lock(&transaction->t_journal->j_list_lock);
+	__journal_file_buffer(jh, transaction, jlist);
+	spin_unlock(&transaction->t_journal->j_list_lock);
+	jbd_unlock_bh_state(jh2bh(jh));
+}
+
+/* 
+ * Remove a buffer from its current buffer list in preparation for
+ * dropping it from its current transaction entirely.  If the buffer has
+ * already started to be used by a subsequent transaction, refile the
+ * buffer on that transaction's metadata list.
+ *
+ * Called under journal->j_list_lock
+ *
+ * Called under jbd_lock_bh_state(jh2bh(jh))
+ */
+void __journal_refile_buffer(struct journal_head *jh)
+{
+	int was_dirty;
+	struct buffer_head *bh = jh2bh(jh);
+
+	J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh));
+	if (jh->b_transaction)
+		assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock);
+
+	/* If the buffer is now unused, just drop it. */
+	if (jh->b_next_transaction == NULL) {
+		__journal_unfile_buffer(jh);
+		return;
+	}
+
+	/*
+	 * It has been modified by a later transaction: add it to the new
+	 * transaction's metadata list.
+	 */
+
+	was_dirty = test_clear_buffer_jbddirty(bh);
+	__journal_temp_unlink_buffer(jh);
+	jh->b_transaction = jh->b_next_transaction;
+	jh->b_next_transaction = NULL;
+	__journal_file_buffer(jh, jh->b_transaction, BJ_Metadata);
+	J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING);
+
+	if (was_dirty)
+		set_buffer_jbddirty(bh);
+}
+
+/*
+ * For the unlocked version of this call, also make sure that any
+ * hanging journal_head is cleaned up if necessary.
+ *
+ * __journal_refile_buffer is usually called as part of a single locked
+ * operation on a buffer_head, in which the caller is probably going to
+ * be hooking the journal_head onto other lists.  In that case it is up
+ * to the caller to remove the journal_head if necessary.  For the
+ * unlocked journal_refile_buffer call, the caller isn't going to be
+ * doing anything else to the buffer so we need to do the cleanup
+ * ourselves to avoid a jh leak. 
+ *
+ * *** The journal_head may be freed by this call! ***
+ */
+void journal_refile_buffer(journal_t *journal, struct journal_head *jh)
+{
+	struct buffer_head *bh = jh2bh(jh);
+
+	jbd_lock_bh_state(bh);
+	spin_lock(&journal->j_list_lock);
+
+	__journal_refile_buffer(jh);
+	jbd_unlock_bh_state(bh);
+	journal_remove_journal_head(bh);
+
+	spin_unlock(&journal->j_list_lock);
+	__brelse(bh);
+}