From 0c2022eccb01630c037f2024531e9ff1afbe1564 Mon Sep 17 00:00:00 2001 From: Yongqiang Yang Date: Mon, 20 Feb 2012 17:53:02 -0500 Subject: jbd2: allocate transaction from separate slab cache There is normally only a handful of these active at any one time, but putting them in a separate slab cache makes debugging memory corruption problems easier. Manish Katiyar also wanted this make it easier to test memory failure scenarios in the jbd2 layer. Cc: Manish Katiyar Signed-off-by: Yongqiang Yang Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 5069b8475150..8adc5d460f56 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -1048,7 +1048,7 @@ restart_loop: jbd_debug(1, "JBD2: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); if (to_free) - kfree(commit_transaction); + jbd2_journal_free_transaction(commit_transaction); wake_up(&journal->j_wait_done_commit); } -- cgit v1.2.3 From 24bcc89c7e7c64982e6192b4952a0a92379fc341 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 13 Mar 2012 15:41:04 -0400 Subject: jbd2: split updating of journal superblock and marking journal empty There are three case of updating journal superblock. In the first case, we want to mark journal as empty (setting s_sequence to 0), in the second case we want to update log tail, in the third case we want to update s_errno. Split these cases into separate functions. It makes the code slightly more straightforward and later patches will make the distinction even more important. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 2 +- fs/jbd2/commit.c | 2 +- fs/jbd2/journal.c | 163 ++++++++++++++++++++++++++------------------ include/linux/jbd2.h | 2 +- include/trace/events/jbd2.h | 12 ++-- 5 files changed, 102 insertions(+), 79 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 253e91890e71..19dcd0b86bca 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -550,7 +550,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal) (journal->j_flags & JBD2_BARRIER)) blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); if (!(journal->j_flags & JBD2_ABORT)) - jbd2_journal_update_superblock(journal, 1); + jbd2_journal_update_sb_log_tail(journal); return 0; } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 8adc5d460f56..19371a8a9015 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -340,7 +340,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) /* Do we need to erase the effects of a prior jbd2_journal_flush? */ if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); - jbd2_journal_update_superblock(journal, 1); + jbd2_journal_update_sb_log_tail(journal); } else { jbd_debug(3, "superblock not updated\n"); } diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index cfb36d99f7a4..6e75fbd75bad 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1110,40 +1110,30 @@ static int journal_reset(journal_t *journal) journal->j_max_transaction_buffers = journal->j_maxlen / 4; - /* Add the dynamic fields and write it to disk. */ - jbd2_journal_update_superblock(journal, 1); - return jbd2_journal_start_thread(journal); -} - -/** - * void jbd2_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 jbd2_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 JBD2_FLUSHED. This avoids + * no recovery (s_start == 0), then we can safely defer the superblock + * update until the next commit by setting JBD2_FLUSHED. This avoids * attempting a write to a potential-readonly device. */ - if (sb->s_start == 0 && journal->j_tail_sequence == - journal->j_transaction_sequence) { + if (sb->s_start == 0) { jbd_debug(1, "JBD2: 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; + journal->j_flags |= JBD2_FLUSHED; + } else { + /* Add the dynamic fields and write it to disk. */ + jbd2_journal_update_sb_log_tail(journal); } + return jbd2_journal_start_thread(journal); +} +static void jbd2_write_superblock(journal_t *journal) +{ + struct buffer_head *bh = journal->j_sb_buffer; + + trace_jbd2_write_superblock(journal); if (buffer_write_io_error(bh)) { /* * Oh, dear. A previous attempt to write the journal @@ -1160,49 +1150,98 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait) set_buffer_uptodate(bh); } + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + sync_dirty_buffer(bh); + if (buffer_write_io_error(bh)) { + printk(KERN_ERR "JBD2: I/O error detected " + "when updating journal superblock for %s.\n", + journal->j_devname); + clear_buffer_write_io_error(bh); + set_buffer_uptodate(bh); + } +} + +/** + * jbd2_journal_update_sb_log_tail() - Update log tail in journal sb on disk. + * @journal: The journal to update. + * + * Update a journal's superblock information about log tail and write it to + * disk, waiting for the IO to complete. + */ +void jbd2_journal_update_sb_log_tail(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + read_lock(&journal->j_state_lock); - jbd_debug(1, "JBD2: updating superblock (start %ld, seq %d, errno %d)\n", - journal->j_tail, journal->j_tail_sequence, journal->j_errno); + jbd_debug(1, "JBD2: updating superblock (start %ld, seq %d)\n", + journal->j_tail, journal->j_tail_sequence); 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); read_unlock(&journal->j_state_lock); - BUFFER_TRACE(bh, "marking dirty"); - mark_buffer_dirty(bh); - if (wait) { - sync_dirty_buffer(bh); - if (buffer_write_io_error(bh)) { - printk(KERN_ERR "JBD2: I/O error detected " - "when updating journal superblock for %s.\n", - journal->j_devname); - clear_buffer_write_io_error(bh); - set_buffer_uptodate(bh); - } - } else - write_dirty_buffer(bh, WRITE); + jbd2_write_superblock(journal); - trace_jbd2_update_superblock_end(journal, wait); + /* Log is no longer empty */ + write_lock(&journal->j_state_lock); + WARN_ON(!sb->s_sequence); + journal->j_flags &= ~JBD2_FLUSHED; + write_unlock(&journal->j_state_lock); +} -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. */ +/** + * jbd2_mark_journal_empty() - Mark on disk journal as empty. + * @journal: The journal to update. + * + * Update a journal's dynamic superblock fields to show that journal is empty. + * Write updated superblock to disk waiting for IO to complete. + */ +static void jbd2_mark_journal_empty(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + + read_lock(&journal->j_state_lock); + jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n", + journal->j_tail_sequence); + + sb->s_sequence = cpu_to_be32(journal->j_tail_sequence); + sb->s_start = cpu_to_be32(0); + read_unlock(&journal->j_state_lock); + + jbd2_write_superblock(journal); + /* Log is no longer empty */ write_lock(&journal->j_state_lock); - if (sb->s_start) - journal->j_flags &= ~JBD2_FLUSHED; - else - journal->j_flags |= JBD2_FLUSHED; + journal->j_flags |= JBD2_FLUSHED; write_unlock(&journal->j_state_lock); } + +/** + * jbd2_journal_update_sb_errno() - Update error in the journal. + * @journal: The journal to update. + * + * Update a journal's errno. Write updated superblock to disk waiting for IO + * to complete. + */ +static void jbd2_journal_update_sb_errno(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + + read_lock(&journal->j_state_lock); + jbd_debug(1, "JBD2: updating superblock error (errno %d)\n", + journal->j_errno); + sb->s_errno = cpu_to_be32(journal->j_errno); + read_unlock(&journal->j_state_lock); + + jbd2_write_superblock(journal); +} + /* * 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; @@ -1395,15 +1434,10 @@ int jbd2_journal_destroy(journal_t *journal) spin_unlock(&journal->j_list_lock); if (journal->j_sb_buffer) { - if (!is_journal_aborted(journal)) { - /* We can now mark the journal as empty. */ - journal->j_tail = 0; - journal->j_tail_sequence = - ++journal->j_transaction_sequence; - jbd2_journal_update_superblock(journal, 1); - } else { + if (!is_journal_aborted(journal)) + jbd2_mark_journal_empty(journal); + else err = -EIO; - } brelse(journal->j_sb_buffer); } @@ -1562,7 +1596,6 @@ int jbd2_journal_flush(journal_t *journal) { int err = 0; transaction_t *transaction = NULL; - unsigned long old_tail; write_lock(&journal->j_state_lock); @@ -1604,14 +1637,8 @@ int jbd2_journal_flush(journal_t *journal) * the magic code for a fully-recovered superblock. Any future * commits of data to the journal will restore the current * s_start value. */ + jbd2_mark_journal_empty(journal); write_lock(&journal->j_state_lock); - old_tail = journal->j_tail; - journal->j_tail = 0; - write_unlock(&journal->j_state_lock); - jbd2_journal_update_superblock(journal, 1); - write_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); @@ -1652,7 +1679,7 @@ int jbd2_journal_wipe(journal_t *journal, int write) err = jbd2_journal_skip_recovery(journal); if (write) - jbd2_journal_update_superblock(journal, 1); + jbd2_mark_journal_empty(journal); no_recovery: return err; @@ -1702,7 +1729,7 @@ static void __journal_abort_soft (journal_t *journal, int errno) __jbd2_journal_abort_hard(journal); if (errno) - jbd2_journal_update_superblock(journal, 1); + jbd2_journal_update_sb_errno(journal); } /** diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 46eef77e6ab8..5f05c77438e5 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -1087,7 +1087,7 @@ extern int jbd2_journal_destroy (journal_t *); extern int jbd2_journal_recover (journal_t *journal); extern int jbd2_journal_wipe (journal_t *, int); extern int jbd2_journal_skip_recovery (journal_t *); -extern void jbd2_journal_update_superblock (journal_t *, int); +extern void jbd2_journal_update_sb_log_tail (journal_t *); extern void __jbd2_journal_abort_hard (journal_t *); extern void jbd2_journal_abort (journal_t *, int); extern int jbd2_journal_errno (journal_t *); diff --git a/include/trace/events/jbd2.h b/include/trace/events/jbd2.h index ae59bc207d70..e05a362bf3f1 100644 --- a/include/trace/events/jbd2.h +++ b/include/trace/events/jbd2.h @@ -236,25 +236,21 @@ TRACE_EVENT(jbd2_cleanup_journal_tail, __entry->block_nr, __entry->freed) ); -TRACE_EVENT(jbd2_update_superblock_end, +TRACE_EVENT(jbd2_write_superblock, - TP_PROTO(journal_t *journal, int wait), + TP_PROTO(journal_t *journal), - TP_ARGS(journal, wait), + TP_ARGS(journal), TP_STRUCT__entry( __field( dev_t, dev ) - __field( int, wait ) ), TP_fast_assign( __entry->dev = journal->j_fs_dev->bd_dev; - __entry->wait = wait; ), - TP_printk("dev %d,%d wait %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->wait) + TP_printk("dev %d,%d", MAJOR(__entry->dev), MINOR(__entry->dev)) ); #endif /* _TRACE_JBD2_H */ -- cgit v1.2.3 From a78bb11d7acd525623c6a0c2ff4e213d527573fa Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 13 Mar 2012 15:43:04 -0400 Subject: jbd2: protect all log tail updates with j_checkpoint_mutex There are some log tail updates that are not protected by j_checkpoint_mutex. Some of these are harmless because they happen during startup or shutdown but updates in jbd2_journal_commit_transaction() and jbd2_journal_flush() can really race with other log tail updates (e.g. someone doing jbd2_journal_flush() with someone running jbd2_cleanup_journal_tail()). So protect all log tail updates with j_checkpoint_mutex. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 2 ++ fs/jbd2/journal.c | 19 ++++++++++++++++--- 2 files changed, 18 insertions(+), 3 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 19371a8a9015..6705717d9b7f 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -340,7 +340,9 @@ void jbd2_journal_commit_transaction(journal_t *journal) /* Do we need to erase the effects of a prior jbd2_journal_flush? */ if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); + mutex_lock(&journal->j_checkpoint_mutex); jbd2_journal_update_sb_log_tail(journal); + mutex_unlock(&journal->j_checkpoint_mutex); } else { jbd_debug(3, "superblock not updated\n"); } diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 6e75fbd75bad..fc5f2acc9f18 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1123,8 +1123,11 @@ static int journal_reset(journal_t *journal) journal->j_errno); journal->j_flags |= JBD2_FLUSHED; } else { + /* Lock here to make assertions happy... */ + mutex_lock(&journal->j_checkpoint_mutex); /* Add the dynamic fields and write it to disk. */ jbd2_journal_update_sb_log_tail(journal); + mutex_unlock(&journal->j_checkpoint_mutex); } return jbd2_journal_start_thread(journal); } @@ -1173,6 +1176,7 @@ void jbd2_journal_update_sb_log_tail(journal_t *journal) { journal_superblock_t *sb = journal->j_superblock; + BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex)); read_lock(&journal->j_state_lock); jbd_debug(1, "JBD2: updating superblock (start %ld, seq %d)\n", journal->j_tail, journal->j_tail_sequence); @@ -1201,6 +1205,7 @@ static void jbd2_mark_journal_empty(journal_t *journal) { journal_superblock_t *sb = journal->j_superblock; + BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex)); read_lock(&journal->j_state_lock); jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n", journal->j_tail_sequence); @@ -1434,9 +1439,11 @@ int jbd2_journal_destroy(journal_t *journal) spin_unlock(&journal->j_list_lock); if (journal->j_sb_buffer) { - if (!is_journal_aborted(journal)) + if (!is_journal_aborted(journal)) { + mutex_lock(&journal->j_checkpoint_mutex); jbd2_mark_journal_empty(journal); - else + mutex_unlock(&journal->j_checkpoint_mutex); + } else err = -EIO; brelse(journal->j_sb_buffer); } @@ -1630,6 +1637,7 @@ int jbd2_journal_flush(journal_t *journal) if (is_journal_aborted(journal)) return -EIO; + mutex_lock(&journal->j_checkpoint_mutex); jbd2_cleanup_journal_tail(journal); /* Finally, mark the journal as really needing no recovery. @@ -1638,6 +1646,7 @@ int jbd2_journal_flush(journal_t *journal) * commits of data to the journal will restore the current * s_start value. */ jbd2_mark_journal_empty(journal); + mutex_unlock(&journal->j_checkpoint_mutex); write_lock(&journal->j_state_lock); J_ASSERT(!journal->j_running_transaction); J_ASSERT(!journal->j_committing_transaction); @@ -1678,8 +1687,12 @@ int jbd2_journal_wipe(journal_t *journal, int write) write ? "Clearing" : "Ignoring"); err = jbd2_journal_skip_recovery(journal); - if (write) + if (write) { + /* Lock to make assertions happy... */ + mutex_lock(&journal->j_checkpoint_mutex); jbd2_mark_journal_empty(journal); + mutex_unlock(&journal->j_checkpoint_mutex); + } no_recovery: return err; -- cgit v1.2.3 From 79feb521a44705262d15cc819a4117a447b11ea7 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 13 Mar 2012 22:22:54 -0400 Subject: jbd2: issue cache flush after checkpointing even with internal journal When we reach jbd2_cleanup_journal_tail(), there is no guarantee that checkpointed buffers are on a stable storage - especially if buffers were written out by jbd2_log_do_checkpoint(), they are likely to be only in disk's caches. Thus when we update journal superblock effectively removing old transaction from journal, this write of superblock can get to stable storage before those checkpointed buffers which can result in filesystem corruption after a crash. Thus we must unconditionally issue a cache flush before we update journal superblock in these cases. A similar problem can also occur if journal superblock is written only in disk's caches, other transaction starts reusing space of the transaction cleaned from the log and power failure happens. Subsequent journal replay would still try to replay the old transaction but some of it's blocks may be already overwritten by the new transaction. For this reason we must use WRITE_FUA when updating log tail and we must first write new log tail to disk and update in-memory information only after that. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/checkpoint.c | 75 ++++-------------------- fs/jbd2/commit.c | 11 +++- fs/jbd2/journal.c | 138 +++++++++++++++++++++++++++++++++++++------- fs/jbd2/recovery.c | 5 +- include/linux/jbd2.h | 6 +- include/trace/events/jbd2.h | 11 ++-- 6 files changed, 155 insertions(+), 91 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 19dcd0b86bca..7f7ee5b90402 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -478,79 +478,28 @@ out: int jbd2_cleanup_journal_tail(journal_t *journal) { - transaction_t * transaction; tid_t first_tid; - unsigned long blocknr, freed; + unsigned long blocknr; if (is_journal_aborted(journal)) return 1; - /* 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. */ - - write_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) { - write_unlock(&journal->j_state_lock); + if (!jbd2_journal_get_log_tail(journal, &first_tid, &blocknr)) 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; - - trace_jbd2_cleanup_journal_tail(journal, first_tid, blocknr, freed); - 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; - write_unlock(&journal->j_state_lock); + J_ASSERT(blocknr != 0); /* - * If there is an external journal, we need to make sure that - * any data blocks that were recently written out --- perhaps - * by jbd2_log_do_checkpoint() --- are flushed out before we - * drop the transactions from the external journal. It's - * unlikely this will be necessary, especially with a - * appropriately sized journal, but we need this to guarantee - * correctness. Fortunately jbd2_cleanup_journal_tail() - * doesn't get called all that often. + * We need to make sure that any blocks that were recently written out + * --- perhaps by jbd2_log_do_checkpoint() --- are flushed out before + * we drop the transactions from the journal. It's unlikely this will + * be necessary, especially with an appropriately sized journal, but we + * need this to guarantee correctness. Fortunately + * jbd2_cleanup_journal_tail() doesn't get called all that often. */ - if ((journal->j_fs_dev != journal->j_dev) && - (journal->j_flags & JBD2_BARRIER)) + if (journal->j_flags & JBD2_BARRIER) blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); - if (!(journal->j_flags & JBD2_ABORT)) - jbd2_journal_update_sb_log_tail(journal); + + __jbd2_update_log_tail(journal, first_tid, blocknr); return 0; } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6705717d9b7f..b89ef84786a7 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -341,7 +341,16 @@ void jbd2_journal_commit_transaction(journal_t *journal) if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); mutex_lock(&journal->j_checkpoint_mutex); - jbd2_journal_update_sb_log_tail(journal); + /* + * We hold j_checkpoint_mutex so tail cannot change under us. + * We don't need any special data guarantees for writing sb + * since journal is empty and it is ok for write to be + * flushed only with transaction commit. + */ + jbd2_journal_update_sb_log_tail(journal, + journal->j_tail_sequence, + journal->j_tail, + WRITE_SYNC); mutex_unlock(&journal->j_checkpoint_mutex); } else { jbd_debug(3, "superblock not updated\n"); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index fc5f2acc9f18..c5ff177400ff 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -742,6 +742,85 @@ struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) return jbd2_journal_add_journal_head(bh); } +/* + * Return tid of the oldest transaction in the journal and block in the journal + * where the transaction starts. + * + * If the journal is now empty, return which will be the next transaction ID + * we will write and where will that transaction start. + * + * The return value is 0 if journal tail cannot be pushed any further, 1 if + * it can. + */ +int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid, + unsigned long *block) +{ + transaction_t *transaction; + int ret; + + read_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + transaction = journal->j_checkpoint_transactions; + if (transaction) { + *tid = transaction->t_tid; + *block = transaction->t_log_start; + } else if ((transaction = journal->j_committing_transaction) != NULL) { + *tid = transaction->t_tid; + *block = transaction->t_log_start; + } else if ((transaction = journal->j_running_transaction) != NULL) { + *tid = transaction->t_tid; + *block = journal->j_head; + } else { + *tid = journal->j_transaction_sequence; + *block = journal->j_head; + } + ret = tid_gt(*tid, journal->j_tail_sequence); + spin_unlock(&journal->j_list_lock); + read_unlock(&journal->j_state_lock); + + return ret; +} + +/* + * Update information in journal structure and in on disk journal superblock + * about log tail. This function does not check whether information passed in + * really pushes log tail further. It's responsibility of the caller to make + * sure provided log tail information is valid (e.g. by holding + * j_checkpoint_mutex all the time between computing log tail and calling this + * function as is the case with jbd2_cleanup_journal_tail()). + * + * Requires j_checkpoint_mutex + */ +void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block) +{ + unsigned long freed; + + BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex)); + + /* + * We cannot afford for write to remain in drive's caches since as + * soon as we update j_tail, next transaction can start reusing journal + * space and if we lose sb update during power failure we'd replay + * old transaction with possibly newly overwritten data. + */ + jbd2_journal_update_sb_log_tail(journal, tid, block, WRITE_FUA); + write_lock(&journal->j_state_lock); + freed = block - journal->j_tail; + if (block < journal->j_tail) + freed += journal->j_last - journal->j_first; + + trace_jbd2_update_log_tail(journal, tid, block, freed); + jbd_debug(1, + "Cleaning journal tail from %d to %d (offset %lu), " + "freeing %lu\n", + journal->j_tail_sequence, tid, block, freed); + + journal->j_free += freed; + journal->j_tail_sequence = tid; + journal->j_tail = block; + write_unlock(&journal->j_state_lock); +} + struct jbd2_stats_proc_session { journal_t *journal; struct transaction_stats_s *stats; @@ -1125,18 +1204,30 @@ static int journal_reset(journal_t *journal) } else { /* Lock here to make assertions happy... */ mutex_lock(&journal->j_checkpoint_mutex); - /* Add the dynamic fields and write it to disk. */ - jbd2_journal_update_sb_log_tail(journal); + /* + * Update log tail information. We use WRITE_FUA since new + * transaction will start reusing journal space and so we + * must make sure information about current log tail is on + * disk before that. + */ + jbd2_journal_update_sb_log_tail(journal, + journal->j_tail_sequence, + journal->j_tail, + WRITE_FUA); mutex_unlock(&journal->j_checkpoint_mutex); } return jbd2_journal_start_thread(journal); } -static void jbd2_write_superblock(journal_t *journal) +static void jbd2_write_superblock(journal_t *journal, int write_op) { struct buffer_head *bh = journal->j_sb_buffer; + int ret; - trace_jbd2_write_superblock(journal); + trace_jbd2_write_superblock(journal, write_op); + if (!(journal->j_flags & JBD2_BARRIER)) + write_op &= ~(REQ_FUA | REQ_FLUSH); + lock_buffer(bh); if (buffer_write_io_error(bh)) { /* * Oh, dear. A previous attempt to write the journal @@ -1152,40 +1243,45 @@ static void jbd2_write_superblock(journal_t *journal) clear_buffer_write_io_error(bh); set_buffer_uptodate(bh); } - - BUFFER_TRACE(bh, "marking dirty"); - mark_buffer_dirty(bh); - sync_dirty_buffer(bh); + get_bh(bh); + bh->b_end_io = end_buffer_write_sync; + ret = submit_bh(write_op, bh); + wait_on_buffer(bh); if (buffer_write_io_error(bh)) { - printk(KERN_ERR "JBD2: I/O error detected " - "when updating journal superblock for %s.\n", - journal->j_devname); clear_buffer_write_io_error(bh); set_buffer_uptodate(bh); + ret = -EIO; + } + if (ret) { + printk(KERN_ERR "JBD2: Error %d detected when updating " + "journal superblock for %s.\n", ret, + journal->j_devname); } } /** * jbd2_journal_update_sb_log_tail() - Update log tail in journal sb on disk. * @journal: The journal to update. + * @tail_tid: TID of the new transaction at the tail of the log + * @tail_block: The first block of the transaction at the tail of the log + * @write_op: With which operation should we write the journal sb * * Update a journal's superblock information about log tail and write it to * disk, waiting for the IO to complete. */ -void jbd2_journal_update_sb_log_tail(journal_t *journal) +void jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid, + unsigned long tail_block, int write_op) { journal_superblock_t *sb = journal->j_superblock; BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex)); - read_lock(&journal->j_state_lock); - jbd_debug(1, "JBD2: updating superblock (start %ld, seq %d)\n", - journal->j_tail, journal->j_tail_sequence); + jbd_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n", + tail_block, tail_tid); - sb->s_sequence = cpu_to_be32(journal->j_tail_sequence); - sb->s_start = cpu_to_be32(journal->j_tail); - read_unlock(&journal->j_state_lock); + sb->s_sequence = cpu_to_be32(tail_tid); + sb->s_start = cpu_to_be32(tail_block); - jbd2_write_superblock(journal); + jbd2_write_superblock(journal, write_op); /* Log is no longer empty */ write_lock(&journal->j_state_lock); @@ -1214,7 +1310,7 @@ static void jbd2_mark_journal_empty(journal_t *journal) sb->s_start = cpu_to_be32(0); read_unlock(&journal->j_state_lock); - jbd2_write_superblock(journal); + jbd2_write_superblock(journal, WRITE_FUA); /* Log is no longer empty */ write_lock(&journal->j_state_lock); @@ -1240,7 +1336,7 @@ static void jbd2_journal_update_sb_errno(journal_t *journal) sb->s_errno = cpu_to_be32(journal->j_errno); read_unlock(&journal->j_state_lock); - jbd2_write_superblock(journal); + jbd2_write_superblock(journal, WRITE_SYNC); } /* diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index da6d7baf1390..c1a03354a22f 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -21,6 +21,7 @@ #include #include #include +#include #endif /* @@ -265,7 +266,9 @@ int jbd2_journal_recover(journal_t *journal) err2 = sync_blockdev(journal->j_fs_dev); if (!err) err = err2; - + /* Make sure all replayed data is on permanent storage */ + if (journal->j_flags & JBD2_BARRIER) + blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); return err; } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 5f05c77438e5..876a7d87192b 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -971,6 +971,9 @@ extern void __journal_clean_data_list(transaction_t *transaction); /* Log buffer allocation */ extern struct journal_head * jbd2_journal_get_descriptor_buffer(journal_t *); int jbd2_journal_next_log_block(journal_t *, unsigned long long *); +int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid, + unsigned long *block); +void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block); /* Commit management */ extern void jbd2_journal_commit_transaction(journal_t *); @@ -1087,7 +1090,8 @@ extern int jbd2_journal_destroy (journal_t *); extern int jbd2_journal_recover (journal_t *journal); extern int jbd2_journal_wipe (journal_t *, int); extern int jbd2_journal_skip_recovery (journal_t *); -extern void jbd2_journal_update_sb_log_tail (journal_t *); +extern void jbd2_journal_update_sb_log_tail (journal_t *, tid_t, + unsigned long, int); extern void __jbd2_journal_abort_hard (journal_t *); extern void jbd2_journal_abort (journal_t *, int); extern int jbd2_journal_errno (journal_t *); diff --git a/include/trace/events/jbd2.h b/include/trace/events/jbd2.h index e05a362bf3f1..127993dbf322 100644 --- a/include/trace/events/jbd2.h +++ b/include/trace/events/jbd2.h @@ -207,7 +207,7 @@ TRACE_EVENT(jbd2_checkpoint_stats, __entry->forced_to_close, __entry->written, __entry->dropped) ); -TRACE_EVENT(jbd2_cleanup_journal_tail, +TRACE_EVENT(jbd2_update_log_tail, TP_PROTO(journal_t *journal, tid_t first_tid, unsigned long block_nr, unsigned long freed), @@ -238,19 +238,22 @@ TRACE_EVENT(jbd2_cleanup_journal_tail, TRACE_EVENT(jbd2_write_superblock, - TP_PROTO(journal_t *journal), + TP_PROTO(journal_t *journal, int write_op), - TP_ARGS(journal), + TP_ARGS(journal, write_op), TP_STRUCT__entry( __field( dev_t, dev ) + __field( int, write_op ) ), TP_fast_assign( __entry->dev = journal->j_fs_dev->bd_dev; + __entry->write_op = write_op; ), - TP_printk("dev %d,%d", MAJOR(__entry->dev), MINOR(__entry->dev)) + TP_printk("dev %d,%d write_op %x", MAJOR(__entry->dev), + MINOR(__entry->dev), __entry->write_op) ); #endif /* _TRACE_JBD2_H */ -- cgit v1.2.3 From 3339578f05787259917788f461f4196b7349c2a4 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 13 Mar 2012 22:45:38 -0400 Subject: jbd2: cleanup journal tail after transaction commit Normally, we have to issue a cache flush before we can update journal tail in journal superblock, effectively wiping out old transactions from the journal. So use the fact that during transaction commit we issue cache flush anyway and opportunistically push journal tail as far as we can. Since update of journal superblock is still costly (we have to use WRITE_FUA), we update log tail only if we can free significant amount of space. Signed-off-by: Jan Kara Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 32 ++++++++++++++++++++++++++++++++ fs/jbd2/journal.c | 13 +++++++++++++ include/linux/jbd2.h | 1 + 3 files changed, 46 insertions(+) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index b89ef84786a7..1dfcb207ea69 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -331,6 +331,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) struct buffer_head *cbh = NULL; /* For transactional checksums */ __u32 crc32_sum = ~0; struct blk_plug plug; + /* Tail of the journal */ + unsigned long first_block; + tid_t first_tid; + int update_tail; /* * First job: lock down the current transaction and wait for @@ -688,10 +692,30 @@ start_journal_io: err = 0; } + /* + * Get current oldest transaction in the log before we issue flush + * to the filesystem device. After the flush we can be sure that + * blocks of all older transactions are checkpointed to persistent + * storage and we will be safe to update journal start in the + * superblock with the numbers we get here. + */ + update_tail = + jbd2_journal_get_log_tail(journal, &first_tid, &first_block); + write_lock(&journal->j_state_lock); + if (update_tail) { + long freed = first_block - journal->j_tail; + + if (first_block < journal->j_tail) + freed += journal->j_last - journal->j_first; + /* Update tail only if we free significant amount of space */ + if (freed < journal->j_maxlen / 4) + update_tail = 0; + } J_ASSERT(commit_transaction->t_state == T_COMMIT); commit_transaction->t_state = T_COMMIT_DFLUSH; write_unlock(&journal->j_state_lock); + /* * If the journal is not located on the file system device, * then we must flush the file system device before we issue @@ -842,6 +866,14 @@ wait_for_iobuf: if (err) jbd2_journal_abort(journal, err); + /* + * Now disk caches for filesystem device are flushed so we are safe to + * erase checkpointed transactions from the log by updating journal + * superblock. + */ + if (update_tail) + jbd2_update_log_tail(journal, first_tid, first_block); + /* 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 diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c5ff177400ff..bda564f63864 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -821,6 +821,19 @@ void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block) write_unlock(&journal->j_state_lock); } +/* + * This is a variaon of __jbd2_update_log_tail which checks for validity of + * provided log tail and locks j_checkpoint_mutex. So it is safe against races + * with other threads updating log tail. + */ +void jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block) +{ + mutex_lock(&journal->j_checkpoint_mutex); + if (tid_gt(tid, journal->j_tail_sequence)) + __jbd2_update_log_tail(journal, tid, block); + mutex_unlock(&journal->j_checkpoint_mutex); +} + struct jbd2_stats_proc_session { journal_t *journal; struct transaction_stats_s *stats; diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 876a7d87192b..912c30a8ddb1 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -974,6 +974,7 @@ int jbd2_journal_next_log_block(journal_t *, unsigned long long *); int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid, unsigned long *block); void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block); +void jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block); /* Commit management */ extern void jbd2_journal_commit_transaction(journal_t *); -- cgit v1.2.3