transaction.c

嵌入式系统设计与实例开发实验教材二源码 多线程应用程序设计 串行端口程序设计 AD接口实验 CAN总线通信实验 GPS通信实验 Linux内核移植与编译实验 IC卡读写实验 SD驱动使

	result = 0;

	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
	
error_out:
	unlock_journal (journal);
	return result;
}


/*
 * journal_restart: 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);

	transaction->t_outstanding_credits -= handle->h_buffer_credits;
	transaction->t_updates--;

	if (!transaction->t_updates)
		wake_up(&journal->j_wait_updates);

	jbd_debug(2, "restarting handle %p\n", handle);
	log_start_commit(journal, transaction);

	handle->h_buffer_credits = nblocks;
	ret = start_this_handle(journal, handle);
	return ret;
}


/* 
 * Barrier operation: establish a transaction barrier. 
 *
 * 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)
{
	lock_journal(journal);
	++journal->j_barrier_count;

	/* Wait until there are no running updates */
	while (1) {
		transaction_t *transaction = journal->j_running_transaction;
		if (!transaction)
			break;
		if (!transaction->t_updates)
			break;
		
		unlock_journal(journal);
		sleep_on(&journal->j_wait_updates);
		lock_journal(journal);
	}

	unlock_journal(journal);

	/* 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);
}

/*
 * Release a transaction barrier obtained with journal_lock_updates().
 *
 * Should be called without the journal lock held.
 */

void journal_unlock_updates (journal_t *journal)
{
	lock_journal(journal);

	J_ASSERT (journal->j_barrier_count != 0);
	
	up(&journal->j_barrier);
	--journal->j_barrier_count;
	wake_up(&journal->j_wait_transaction_locked);
	unlock_journal(journal);
}

/*
 * journal_get_write_access: notify intent to modify a buffer for metadata
 * (not data) update.
 *
 * 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).
 *
 * 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.
 */

static int
do_get_write_access(handle_t *handle, struct journal_head *jh, int force_copy) 
{
	transaction_t *transaction = handle->h_transaction;
	journal_t *journal = transaction->t_journal;
	int error;
	char *frozen_buffer = NULL;
	int need_copy = 0;

	jbd_debug(5, "buffer_head %p, force_copy %d\n", jh, force_copy);

	JBUFFER_TRACE(jh, "entry");
repeat:
	/* @@@ Need to check for errors here at some point. */

	/*
	 * AKPM: neither bdflush nor kupdate run with the BKL.   There's
	 * nothing we can do to prevent them from starting writeout of a
	 * BUF_DIRTY buffer at any time.  And checkpointing buffers are on
	 * BUF_DIRTY.  So.  We no longer assert that the buffer is unlocked.
	 *
	 * However.  It is very wrong for us to allow ext3 to start directly
	 * altering the ->b_data of buffers which may at that very time be
	 * undergoing writeout to the client filesystem.  This can leave
	 * the filesystem in an inconsistent, transient state if we crash.
	 * So what we do is to steal the buffer if it is in checkpoint
	 * mode and dirty.  The journal lock will keep out checkpoint-mode
	 * state transitions within journal_remove_checkpoint() and the buffer
	 * is locked to keep bdflush/kupdate/whoever away from it as well.
	 *
	 * AKPM: we have replaced all the lock_journal_bh_wait() stuff with a
	 * simple lock_journal().  This code here will care for locked buffers.
	 */
	/*
	 * The buffer_locked() || buffer_dirty() tests here are simply an
	 * optimisation tweak.  If anyone else in the system decides to
	 * lock this buffer later on, we'll blow up.  There doesn't seem
	 * to be a good reason why they should do this.
	 */
	if (jh->b_cp_transaction &&
	    (buffer_locked(jh2bh(jh)) || buffer_dirty(jh2bh(jh)))) {
		unlock_journal(journal);
		lock_buffer(jh2bh(jh));
		spin_lock(&journal_datalist_lock);
		if (jh->b_cp_transaction && buffer_dirty(jh2bh(jh))) {
			/* OK, we need to steal it */
			JBUFFER_TRACE(jh, "stealing from checkpoint mode");
			J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
			J_ASSERT_JH(jh, jh->b_frozen_data == NULL);

			J_ASSERT(handle->h_buffer_credits > 0);
			handle->h_buffer_credits--;

			/* This will clear BH_Dirty and set BH_JBDDirty. */
			JBUFFER_TRACE(jh, "file as BJ_Reserved");
			__journal_file_buffer(jh, transaction, BJ_Reserved);

			/* And pull it off BUF_DIRTY, onto BUF_CLEAN */
			refile_buffer(jh2bh(jh));

			/*
			 * The buffer is now hidden from bdflush.   It is
			 * metadata against the current transaction.
			 */
			JBUFFER_TRACE(jh, "steal from cp mode is complete");
		}
		spin_unlock(&journal_datalist_lock);
		unlock_buffer(jh2bh(jh));
		lock_journal(journal);
		goto repeat;
	}

	J_ASSERT_JH(jh, !buffer_locked(jh2bh(jh)));

	error = -EROFS;
	if (is_handle_aborted(handle)) 
		goto out_unlocked;
	error = 0;

	spin_lock(&journal_datalist_lock);

	/* 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_locked;

	/* 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;

		J_ASSERT_JH(jh, handle->h_buffer_credits > 0);
		handle->h_buffer_credits--;
		goto done_locked;
	}
	
	/* 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) {
			JBUFFER_TRACE(jh, "on shadow: sleep");
			spin_unlock(&journal_datalist_lock);
			unlock_journal(journal);
			/* commit wakes up all shadow buffers after IO */
			sleep_on(&jh2bh(jh)->b_wait);
			lock_journal(journal);
			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");
				spin_unlock(&journal_datalist_lock);
				unlock_journal(journal);
				frozen_buffer = jbd_kmalloc(jh2bh(jh)->b_size,
							    GFP_NOFS);
				lock_journal(journal);
				if (!frozen_buffer) {
					printk(KERN_EMERG __FUNCTION__
						"OOM for frozen_buffer\n");
					JBUFFER_TRACE(jh, "oom!");
					error = -ENOMEM;
					spin_lock(&journal_datalist_lock);
					goto done_locked;
				}
				goto repeat;
			}

			jh->b_frozen_data = frozen_buffer;
			frozen_buffer = NULL;
			need_copy = 1;
		}
		jh->b_next_transaction = transaction;
	}

	J_ASSERT(handle->h_buffer_credits > 0);
	handle->h_buffer_credits--;

	/* 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");
		__journal_file_buffer(jh, transaction, BJ_Reserved);
	}
	
done_locked:
	spin_unlock(&journal_datalist_lock);
	if (need_copy) {
		struct page *page;
		int offset;
		char *source;

		J_ASSERT_JH(jh, buffer_uptodate(jh2bh(jh)));
		page = jh2bh(jh)->b_page;
		offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK;
		source = kmap(page);
		memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size);
		kunmap(page);
	}
	

	/* If we are about to journal a buffer, then any revoke pending
           on it is no longer valid. */
	journal_cancel_revoke(handle, jh);

out_unlocked:
	if (frozen_buffer)
		kfree(frozen_buffer);

	JBUFFER_TRACE(jh, "exit");
	return error;
}

int journal_get_write_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 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. */
	lock_journal(journal);
	rc = do_get_write_access(handle, jh, 0);
	journal_unlock_journal_head(jh);
	unlock_journal(journal);
	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 (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);
	lock_journal(journal);
	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. */
	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)));

	J_ASSERT_JH(jh, handle->h_buffer_credits > 0);
	handle->h_buffer_credits--;

	spin_lock(&journal_datalist_lock);
	if (jh->b_transaction == NULL) {
		jh->b_transaction = transaction;
		JBUFFER_TRACE(jh, "file as BJ_Reserved");
		__journal_file_buffer(jh, transaction, BJ_Reserved);
		JBUFFER_TRACE(jh, "refile");
		refile_buffer(jh2bh(jh));
	} else if (jh->b_transaction == journal->j_committing_transaction) {
		JBUFFER_TRACE(jh, "set next transaction");
		jh->b_next_transaction = transaction;
	}
	spin_unlock(&journal_datalist_lock);