From cc3e1bea5d87635c519da657303690f5538bb4eb Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 23 Dec 2009 06:52:08 -0500 Subject: ext4, jbd2: Add barriers for file systems with exernal journals This is a bit complicated because we are trying to optimize when we send barriers to the fs data disk. We could just throw in an extra barrier to the data disk whenever we send a barrier to the journal disk, but that's not always strictly necessary. We only need to send a barrier during a commit when there are data blocks which are must be written out due to an inode written in ordered mode, or if fsync() depends on the commit to force data blocks to disk. Finally, before we drop transactions from the beginning of the journal during a checkpoint operation, we need to guarantee that any blocks that were flushed out to the data disk are firmly on the rust platter before we drop the transaction from the journal. Thanks to Oleg Drokin for pointing out this flaw in ext3/ext4. Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6a10238d2c63..1bc74b6f26d2 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -259,6 +259,7 @@ static int journal_submit_data_buffers(journal_t *journal, ret = err; spin_lock(&journal->j_list_lock); J_ASSERT(jinode->i_transaction == commit_transaction); + commit_transaction->t_flushed_data_blocks = 1; jinode->i_flags &= ~JI_COMMIT_RUNNING; wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); } @@ -708,8 +709,17 @@ start_journal_io: } } - /* Done it all: now write the commit record asynchronously. */ + /* + * If the journal is not located on the file system device, + * then we must flush the file system device before we issue + * the commit record + */ + if (commit_transaction->t_flushed_data_blocks && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, NULL); + /* Done it all: now write the commit record asynchronously. */ if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { err = journal_submit_commit_record(journal, commit_transaction, @@ -720,13 +730,6 @@ start_journal_io: blkdev_issue_flush(journal->j_dev, NULL); } - /* - * This is the right place to wait for data buffers both for ASYNC - * and !ASYNC commit. If commit is ASYNC, we need to wait only after - * the commit block went to disk (which happens above). If commit is - * SYNC, we need to wait for data buffers before we start writing - * commit block, which happens below in such setting. - */ err = journal_finish_inode_data_buffers(journal, commit_transaction); if (err) { printk(KERN_WARNING -- cgit v1.2.2 From ba869023eac8354b17acdcff82b851ea8e7b1809 Mon Sep 17 00:00:00 2001 From: dingdinghua Date: Mon, 15 Feb 2010 16:35:42 -0500 Subject: jbd2: delay discarding buffers in journal_unmap_buffer Delay discarding buffers in journal_unmap_buffer until we know that "add to orphan" operation has definitely been committed, otherwise the log space of committing transation may be freed and reused before truncate get committed, updates may get lost if crash happens. Signed-off-by: dingdinghua Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 1bc74b6f26d2..3ee211ed58f1 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -930,12 +930,12 @@ restart_loop: /* 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 + * that buffer in the future after the "add to orphan" + * operation 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)) { + if (buffer_freed(bh) && !jh->b_next_transaction) { clear_buffer_freed(bh); clear_buffer_jbddirty(bh); } -- cgit v1.2.2 From 23e2af3518facab6838d7aac1f46fbd7a5a290ce Mon Sep 17 00:00:00 2001 From: dingdinghua Date: Wed, 24 Feb 2010 12:11:20 -0500 Subject: jbd2: clean up an assertion in jbd2_journal_commit_transaction() commit_transaction has the same value as journal->j_running_transaction, so we can simplify the assert statement. Signed-off-by: dingdinghua Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 3ee211ed58f1..671da7fb7ffd 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -883,8 +883,7 @@ restart_loop: spin_unlock(&journal->j_list_lock); bh = jh2bh(jh); jbd_lock_bh_state(bh); - J_ASSERT_JH(jh, jh->b_transaction == commit_transaction || - jh->b_transaction == journal->j_running_transaction); + J_ASSERT_JH(jh, jh->b_transaction == commit_transaction); /* * If there is undo-protected committed data against -- cgit v1.2.2