commit.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 842 行 · 第 1/2 页

/*
 * linux/fs/commit.c
 *
 * Written by Stephen C. Tweedie <[e-mail]>, 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;
}

/*
 * 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[64];
	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");

	/*
	 * 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_unfile_buffer(jh);
			__journal_file_buffer(jh, commit_transaction,
						BJ_Locked);
			jbd_unlock_bh_state(bh);
			if (need_resched()) {
				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 == ARRAY_SIZE(wbuf)) {
					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 (need_resched()) {
					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);
		if (need_resched()) {
			spin_unlock(&journal->j_list_lock);
			cond_resched();
			spin_lock(&journal->j_list_lock);
		}
	}
	spin_unlock(&journal->j_list_lock);

	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");