diff options
Diffstat (limited to 'fs/jbd2/commit.c')
| -rw-r--r-- | fs/jbd2/commit.c | 1034 |
1 files changed, 636 insertions, 398 deletions
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6bd8005e3d34..cf2fc0594063 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -20,24 +20,38 @@ #include <linux/slab.h> #include <linux/mm.h> #include <linux/pagemap.h> -#include <linux/smp_lock.h> +#include <linux/jiffies.h> +#include <linux/crc32.h> +#include <linux/writeback.h> +#include <linux/backing-dev.h> +#include <linux/bio.h> +#include <linux/blkdev.h> +#include <linux/bitops.h> +#include <trace/events/jbd2.h> /* - * Default IO end handler for temporary BJ_IO buffer_heads. + * IO end handler for temporary buffer_heads handling writes to the journal. */ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate) { + struct buffer_head *orig_bh = bh->b_private; + BUFFER_TRACE(bh, ""); if (uptodate) set_buffer_uptodate(bh); else clear_buffer_uptodate(bh); + if (orig_bh) { + clear_bit_unlock(BH_Shadow, &orig_bh->b_state); + smp_mb__after_clear_bit(); + wake_up_bit(&orig_bh->b_state, BH_Shadow); + } 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. + * When an ext4 file is truncated, it is possible that some pages are not + * successfully 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 @@ -64,7 +78,7 @@ static void release_buffer_page(struct buffer_head *bh) goto nope; /* OK, it's a truncated page */ - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto nope; page_cache_get(page); @@ -78,211 +92,272 @@ 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) +static void jbd2_commit_block_csum_set(journal_t *j, struct buffer_head *bh) { - if (!jbd_trylock_bh_state(bh)) { - spin_unlock(&journal->j_list_lock); - schedule(); - return 0; - } - return 1; + struct commit_header *h; + __u32 csum; + + if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + return; + + h = (struct commit_header *)(bh->b_data); + h->h_chksum_type = 0; + h->h_chksum_size = 0; + h->h_chksum[0] = 0; + csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize); + h->h_chksum[0] = cpu_to_be32(csum); } -/* Done it all: now write the commit record. We should have +/* + * Done it all: now submit 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) +static int journal_submit_commit_record(journal_t *journal, + transaction_t *commit_transaction, + struct buffer_head **cbh, + __u32 crc32_sum) { - struct journal_head *descriptor; + struct commit_header *tmp; struct buffer_head *bh; - int i, ret; - int barrier_done = 0; + int ret; + struct timespec now = current_kernel_time(); + + *cbh = NULL; if (is_journal_aborted(journal)) return 0; - descriptor = jbd2_journal_get_descriptor_buffer(journal); - if (!descriptor) + bh = jbd2_journal_get_descriptor_buffer(journal); + if (!bh) 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(JBD2_MAGIC_NUMBER); - tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); - tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + tmp = (struct commit_header *)bh->b_data; + tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); + tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + tmp->h_commit_sec = cpu_to_be64(now.tv_sec); + tmp->h_commit_nsec = cpu_to_be32(now.tv_nsec); + + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + tmp->h_chksum_type = JBD2_CRC32_CHKSUM; + tmp->h_chksum_size = JBD2_CRC32_CHKSUM_SIZE; + tmp->h_chksum[0] = cpu_to_be32(crc32_sum); } + jbd2_commit_block_csum_set(journal, bh); + + BUFFER_TRACE(bh, "submit commit block"); + lock_buffer(bh); + clear_buffer_dirty(bh); + set_buffer_uptodate(bh); + bh->b_end_io = journal_end_buffer_io_sync; + + if (journal->j_flags & JBD2_BARRIER && + !JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) + ret = submit_bh(WRITE_SYNC | WRITE_FLUSH_FUA, bh); + else + ret = submit_bh(WRITE_SYNC, bh); - JBUFFER_TRACE(descriptor, "write commit block"); - set_buffer_dirty(bh); - if (journal->j_flags & JBD2_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]; + *cbh = bh; + return ret; +} - 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 &= ~JBD2_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() */ - jbd2_journal_put_journal_head(descriptor); +/* + * This function along with journal_submit_commit_record + * allows to write the commit record asynchronously. + */ +static int journal_wait_on_commit_record(journal_t *journal, + struct buffer_head *bh) +{ + int ret = 0; - return (ret == -EIO); + clear_buffer_dirty(bh); + wait_on_buffer(bh); + + if (unlikely(!buffer_uptodate(bh))) + ret = -EIO; + put_bh(bh); /* One for getblk() */ + + return ret; } -static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) +/* + * write the filemap data using writepage() address_space_operations. + * We don't do block allocation here even for delalloc. We don't + * use writepages() because with dealyed allocation we may be doing + * block allocation in writepages(). + */ +static int journal_submit_inode_data_buffers(struct address_space *mapping) { - int i; - - for (i = 0; i < bufs; i++) { - wbuf[i]->b_end_io = end_buffer_write_sync; - /* We use-up our safety reference in submit_bh() */ - submit_bh(WRITE, wbuf[i]); - } + int ret; + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = mapping->nrpages * 2, + .range_start = 0, + .range_end = i_size_read(mapping->host), + }; + + ret = generic_writepages(mapping, &wbc); + return ret; } /* - * Submit all the data buffers to disk + * Submit all the data buffers of inode associated with the transaction to + * disk. + * + * We are in a committing transaction. Therefore no new inode can be added to + * our inode list. We use JI_COMMIT_RUNNING flag to protect inode we currently + * operate on from being released while we write out pages. */ -static void journal_submit_data_buffers(journal_t *journal, - transaction_t *commit_transaction) +static int journal_submit_data_buffers(journal_t *journal, + transaction_t *commit_transaction) { - struct journal_head *jh; - struct buffer_head *bh; - int locked; - int bufs = 0; - struct buffer_head **wbuf = journal->j_wbuf; + struct jbd2_inode *jinode; + int err, ret = 0; + struct address_space *mapping; - /* - * 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. - * - * 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); + list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) { + mapping = jinode->i_vfs_inode->i_mapping; + set_bit(__JI_COMMIT_RUNNING, &jinode->i_flags); + spin_unlock(&journal->j_list_lock); + /* + * submit the inode data buffers. We use writepage + * instead of writepages. Because writepages can do + * block allocation with delalloc. We need to write + * only allocated blocks here. + */ + trace_jbd2_submit_inode_data(jinode->i_vfs_inode); + err = journal_submit_inode_data_buffers(mapping); + if (!ret) + ret = err; + spin_lock(&journal->j_list_lock); + J_ASSERT(jinode->i_transaction == commit_transaction); + clear_bit(__JI_COMMIT_RUNNING, &jinode->i_flags); + smp_mb__after_clear_bit(); + wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); + } + spin_unlock(&journal->j_list_lock); + return ret; +} - while (commit_transaction->t_sync_datalist) { - jh = commit_transaction->t_sync_datalist; - bh = jh2bh(jh); - locked = 0; +/* + * Wait for data submitted for writeout, refile inodes to proper + * transaction if needed. + * + */ +static int journal_finish_inode_data_buffers(journal_t *journal, + transaction_t *commit_transaction) +{ + struct jbd2_inode *jinode, *next_i; + int err, ret = 0; - /* Get reference just to make sure buffer does not disappear - * when we are forced to drop various locks */ - get_bh(bh); - /* If the buffer is dirty, we need to submit IO and hence - * we need the buffer lock. We try to lock the buffer without - * blocking. If we fail, we need to drop j_list_lock and do - * blocking lock_buffer(). - */ - if (buffer_dirty(bh)) { - if (test_set_buffer_locked(bh)) { - BUFFER_TRACE(bh, "needs blocking lock"); - spin_unlock(&journal->j_list_lock); - /* Write out all data to prevent deadlocks */ - journal_do_submit_data(wbuf, bufs); - bufs = 0; - lock_buffer(bh); - spin_lock(&journal->j_list_lock); - } - locked = 1; - } - /* We have to get bh_state lock. Again out of order, sigh. */ - if (!inverted_lock(journal, bh)) { - jbd_lock_bh_state(bh); - spin_lock(&journal->j_list_lock); - } - /* Someone already cleaned up the buffer? */ - if (!buffer_jbd(bh) - || jh->b_transaction != commit_transaction - || jh->b_jlist != BJ_SyncData) { - jbd_unlock_bh_state(bh); - if (locked) - unlock_buffer(bh); - BUFFER_TRACE(bh, "already cleaned up"); - put_bh(bh); - continue; - } - if (locked && test_clear_buffer_dirty(bh)) { - BUFFER_TRACE(bh, "needs writeout, adding to array"); - wbuf[bufs++] = bh; - __jbd2_journal_file_buffer(jh, commit_transaction, - BJ_Locked); - jbd_unlock_bh_state(bh); - if (bufs == journal->j_wbufsize) { - spin_unlock(&journal->j_list_lock); - journal_do_submit_data(wbuf, bufs); - bufs = 0; - goto write_out_data; - } - } else if (!locked && buffer_locked(bh)) { - __jbd2_journal_file_buffer(jh, commit_transaction, - BJ_Locked); - jbd_unlock_bh_state(bh); - put_bh(bh); - } else { - BUFFER_TRACE(bh, "writeout complete: unfile"); - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - if (locked) - unlock_buffer(bh); - jbd2_journal_remove_journal_head(bh); - /* Once for our safety reference, once for - * jbd2_journal_remove_journal_head() */ - put_bh(bh); - put_bh(bh); + /* For locking, see the comment in journal_submit_data_buffers() */ + spin_lock(&journal->j_list_lock); + list_for_each_entry(jinode, &commit_transaction->t_inode_list, i_list) { + set_bit(__JI_COMMIT_RUNNING, &jinode->i_flags); + spin_unlock(&journal->j_list_lock); + err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping); + if (err) { + /* + * Because AS_EIO is cleared by + * filemap_fdatawait_range(), set it again so + * that user process can get -EIO from fsync(). + */ + set_bit(AS_EIO, + &jinode->i_vfs_inode->i_mapping->flags); + + if (!ret) + ret = err; } + spin_lock(&journal->j_list_lock); + clear_bit(__JI_COMMIT_RUNNING, &jinode->i_flags); + smp_mb__after_clear_bit(); + wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); + } - if (lock_need_resched(&journal->j_list_lock)) { - spin_unlock(&journal->j_list_lock); - goto write_out_data; + /* Now refile inode to proper lists */ + list_for_each_entry_safe(jinode, next_i, + &commit_transaction->t_inode_list, i_list) { + list_del(&jinode->i_list); + if (jinode->i_next_transaction) { + jinode->i_transaction = jinode->i_next_transaction; + jinode->i_next_transaction = NULL; + list_add(&jinode->i_list, + &jinode->i_transaction->t_inode_list); + } else { + jinode->i_transaction = NULL; } } spin_unlock(&journal->j_list_lock); - journal_do_submit_data(wbuf, bufs); + + return ret; +} + +static __u32 jbd2_checksum_data(__u32 crc32_sum, struct buffer_head *bh) +{ + struct page *page = bh->b_page; + char *addr; + __u32 checksum; + + addr = kmap_atomic(page); + checksum = crc32_be(crc32_sum, + (void *)(addr + offset_in_page(bh->b_data)), bh->b_size); + kunmap_atomic(addr); + + return checksum; } -static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, +static void write_tag_block(int tag_bytes, journal_block_tag_t *tag, unsigned long long block) { tag->t_blocknr = cpu_to_be32(block & (u32)~0); - if (tag_bytes > JBD_TAG_SIZE32) + if (tag_bytes > JBD2_TAG_SIZE32) tag->t_blocknr_high = cpu_to_be32((block >> 31) >> 1); } +static void jbd2_descr_block_csum_set(journal_t *j, + struct buffer_head *bh) +{ + struct jbd2_journal_block_tail *tail; + __u32 csum; + + if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + return; + + tail = (struct jbd2_journal_block_tail *)(bh->b_data + j->j_blocksize - + sizeof(struct jbd2_journal_block_tail)); + tail->t_checksum = 0; + csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize); + tail->t_checksum = cpu_to_be32(csum); +} + +static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag, + struct buffer_head *bh, __u32 sequence) +{ + struct page *page = bh->b_page; + __u8 *addr; + __u32 csum32; + __be32 seq; + + if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + return; + + seq = cpu_to_be32(sequence); + addr = kmap_atomic(page); + csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&seq, sizeof(seq)); + csum32 = jbd2_chksum(j, csum32, addr + offset_in_page(bh->b_data), + bh->b_size); + kunmap_atomic(addr); + + /* We only have space to store the lower 16 bits of the crc32c. */ + tag->t_checksum = cpu_to_be16(csum32); +} /* * jbd2_journal_commit_transaction * @@ -291,13 +366,17 @@ static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, */ void jbd2_journal_commit_transaction(journal_t *journal) { + struct transaction_stats_s stats; transaction_t *commit_transaction; - struct journal_head *jh, *new_jh, *descriptor; + struct journal_head *jh; + struct buffer_head *descriptor; struct buffer_head **wbuf = journal->j_wbuf; int bufs; int flags; int err; unsigned long long blocknr; + ktime_t start_time; + u64 commit_time; char *tagp = NULL; journal_header_t *header; journal_block_tag_t *tag = NULL; @@ -306,22 +385,40 @@ void jbd2_journal_commit_transaction(journal_t *journal) int tag_flag; int i; int tag_bytes = journal_tag_bytes(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; + int csum_size = 0; + LIST_HEAD(io_bufs); + LIST_HEAD(log_bufs); + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) + csum_size = sizeof(struct jbd2_journal_block_tail); /* * 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 jbd2_journal_flush? */ if (journal->j_flags & JBD2_FLUSHED) { jbd_debug(3, "super block updated\n"); - jbd2_journal_update_superblock(journal, 1); + mutex_lock(&journal->j_checkpoint_mutex); + /* + * 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"); } @@ -330,32 +427,44 @@ void jbd2_journal_commit_transaction(journal_t *journal) 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", + trace_jbd2_start_commit(journal, commit_transaction); + jbd_debug(1, "JBD2: starting commit of transaction %d\n", commit_transaction->t_tid); - spin_lock(&journal->j_state_lock); + write_lock(&journal->j_state_lock); + J_ASSERT(commit_transaction->t_state == T_RUNNING); commit_transaction->t_state = T_LOCKED; + trace_jbd2_commit_locking(journal, commit_transaction); + stats.run.rs_wait = commit_transaction->t_max_wait; + stats.run.rs_request_delay = 0; + stats.run.rs_locked = jiffies; + if (commit_transaction->t_requested) + stats.run.rs_request_delay = + jbd2_time_diff(commit_transaction->t_requested, + stats.run.rs_locked); + stats.run.rs_running = jbd2_time_diff(commit_transaction->t_start, + stats.run.rs_locked); + spin_lock(&commit_transaction->t_handle_lock); - while (commit_transaction->t_updates) { + while (atomic_read(&commit_transaction->t_updates)) { DEFINE_WAIT(wait); prepare_to_wait(&journal->j_wait_updates, &wait, TASK_UNINTERRUPTIBLE); - if (commit_transaction->t_updates) { + if (atomic_read(&commit_transaction->t_updates)) { spin_unlock(&commit_transaction->t_handle_lock); - spin_unlock(&journal->j_state_lock); + write_unlock(&journal->j_state_lock); schedule(); - spin_lock(&journal->j_state_lock); + write_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 <= + J_ASSERT (atomic_read(&commit_transaction->t_outstanding_credits) <= journal->j_max_transaction_buffers); /* @@ -372,7 +481,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) * we do not require it to remember exactly which old buffers it * has reserved. This is consistent with the existing behaviour * that multiple jbd2_journal_get_write_access() calls to the same - * buffer are perfectly permissable. + * buffer are perfectly permissible. */ while (commit_transaction->t_reserved_list) { jh = commit_transaction->t_reserved_list; @@ -385,7 +494,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) struct buffer_head *bh = jh2bh(jh); jbd_lock_bh_state(bh); - jbd2_slab_free(jh->b_committed_data, bh->b_size); + jbd2_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; jbd_unlock_bh_state(bh); } @@ -401,106 +510,79 @@ void jbd2_journal_commit_transaction(journal_t *journal) __jbd2_journal_clean_checkpoint_list(journal); spin_unlock(&journal->j_list_lock); - jbd_debug (3, "JBD: commit phase 1\n"); + jbd_debug(3, "JBD2: commit phase 1\n"); + + /* + * Clear revoked flag to reflect there is no revoked buffers + * in the next transaction which is going to be started. + */ + jbd2_clear_buffer_revoked_flags(journal); /* * Switch to a new revoke table. */ jbd2_journal_switch_revoke_table(journal); + /* + * Reserved credits cannot be claimed anymore, free them + */ + atomic_sub(atomic_read(&journal->j_reserved_credits), + &commit_transaction->t_outstanding_credits); + + trace_jbd2_commit_flushing(journal, commit_transaction); + stats.run.rs_flushing = jiffies; + stats.run.rs_locked = jbd2_time_diff(stats.run.rs_locked, + stats.run.rs_flushing); + commit_transaction->t_state = T_FLUSH; journal->j_committing_transaction = commit_transaction; journal->j_running_transaction = NULL; + start_time = ktime_get(); commit_transaction->t_log_start = journal->j_head; wake_up(&journal->j_wait_transaction_locked); - spin_unlock(&journal->j_state_lock); + write_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); + jbd_debug(3, "JBD2: commit phase 2a\n"); /* * Now start flushing things to disk, in the order they appear * on the transaction lists. Data blocks go first. */ - err = 0; - journal_submit_data_buffers(journal, commit_transaction); - - /* - * Wait for all previously submitted IO to complete. - */ - spin_lock(&journal->j_list_lock); - 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) { - __jbd2_journal_unfile_buffer(jh); - jbd_unlock_bh_state(bh); - jbd2_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); - + err = journal_submit_data_buffers(journal, commit_transaction); if (err) - __jbd2_journal_abort_hard(journal); - - jbd2_journal_write_revoke_records(journal, commit_transaction); + jbd2_journal_abort(journal, err); - 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); + blk_start_plug(&plug); + jbd2_journal_write_revoke_records(journal, commit_transaction, + &log_bufs, WRITE_SYNC); + blk_finish_plug(&plug); - jbd_debug (3, "JBD: commit phase 3\n"); + jbd_debug(3, "JBD2: commit phase 2b\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: */ + write_lock(&journal->j_state_lock); commit_transaction->t_state = T_COMMIT; + write_unlock(&journal->j_state_lock); - descriptor = NULL; + trace_jbd2_commit_logging(journal, commit_transaction); + stats.run.rs_logging = jiffies; + stats.run.rs_flushing = jbd2_time_diff(stats.run.rs_flushing, + stats.run.rs_logging); + stats.run.rs_blocks = + atomic_read(&commit_transaction->t_outstanding_credits); + stats.run.rs_blocks_logged = 0; + + J_ASSERT(commit_transaction->t_nr_buffers <= + atomic_read(&commit_transaction->t_outstanding_credits)); + + err = 0; bufs = 0; + descriptor = NULL; + blk_start_plug(&plug); while (commit_transaction->t_buffers) { /* Find the next buffer to be journaled... */ @@ -508,10 +590,15 @@ void jbd2_journal_commit_transaction(journal_t *journal) jh = commit_transaction->t_buffers; /* If we're in abort mode, we just un-journal the buffer and - release it for background writing. */ + release it. */ if (is_journal_aborted(journal)) { + clear_buffer_jbddirty(jh2bh(jh)); JBUFFER_TRACE(jh, "journal is aborting: refile"); + jbd2_buffer_abort_trigger(jh, + jh->b_frozen_data ? + jh->b_frozen_triggers : + jh->b_triggers); jbd2_journal_refile_buffer(journal, jh); /* If that was the last one, we need to clean up * any descriptor buffers which may have been @@ -526,38 +613,36 @@ void jbd2_journal_commit_transaction(journal_t *journal) record the metadata buffer. */ if (!descriptor) { - struct buffer_head *bh; - J_ASSERT (bufs == 0); - jbd_debug(4, "JBD: get descriptor\n"); + jbd_debug(4, "JBD2: get descriptor\n"); descriptor = jbd2_journal_get_descriptor_buffer(journal); if (!descriptor) { - __jbd2_journal_abort_hard(journal); + jbd2_journal_abort(journal, -EIO); 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]; + jbd_debug(4, "JBD2: got buffer %llu (%p)\n", + (unsigned long long)descriptor->b_blocknr, + descriptor->b_data); + header = (journal_header_t *)descriptor->b_data; header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); header->h_blocktype = cpu_to_be32(JBD2_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); + tagp = &descriptor->b_data[sizeof(journal_header_t)]; + space_left = descriptor->b_size - + sizeof(journal_header_t); first_tag = 1; - set_buffer_jwrite(bh); - set_buffer_dirty(bh); - wbuf[bufs++] = bh; + set_buffer_jwrite(descriptor); + set_buffer_dirty(descriptor); + wbuf[bufs++] = descriptor; /* Record it so that we can wait for IO completion later */ - BUFFER_TRACE(bh, "ph3: file as descriptor"); - jbd2_journal_file_buffer(descriptor, commit_transaction, - BJ_LogCtl); + BUFFER_TRACE(descriptor, "ph3: file as descriptor"); + jbd2_file_log_bh(&log_bufs, descriptor); } /* Where is the buffer to be written? */ @@ -567,7 +652,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) and repeat this loop: we'll fall into the refile-on-abort condition above. */ if (err) { - __jbd2_journal_abort_hard(journal); + jbd2_journal_abort(journal, err); continue; } @@ -576,29 +661,26 @@ void jbd2_journal_commit_transaction(journal_t *journal) * the free space in the log, but this counter is changed * by jbd2_journal_next_log_block() also. */ - commit_transaction->t_outstanding_credits--; + atomic_dec(&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. */ + rid of the 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: jbd2_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) + * Make a temporary IO buffer with which to write it out + * (this will requeue the metadata buffer to BJ_Shadow). */ + set_bit(BH_JWrite, &jh2bh(jh)->b_state); JBUFFER_TRACE(jh, "ph3: write metadata"); flags = jbd2_journal_write_metadata_buffer(commit_transaction, - jh, &new_jh, blocknr); - set_bit(BH_JWrite, &jh2bh(new_jh)->b_state); - wbuf[bufs++] = jh2bh(new_jh); + jh, &wbuf[bufs], blocknr); + if (flags < 0) { + jbd2_journal_abort(journal, flags); + continue; + } + jbd2_file_log_bh(&io_bufs, wbuf[bufs]); /* Record the new block's tag in the current descriptor buffer */ @@ -611,9 +693,12 @@ void jbd2_journal_commit_transaction(journal_t *journal) tag = (journal_block_tag_t *) tagp; write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr); - tag->t_flags = cpu_to_be32(tag_flag); + tag->t_flags = cpu_to_be16(tag_flag); + jbd2_block_tag_csum_set(journal, tag, wbuf[bufs], + commit_transaction->t_tid); tagp += tag_bytes; space_left -= tag_bytes; + bufs++; if (first_tag) { memcpy (tagp, journal->j_uuid, 16); @@ -627,26 +712,37 @@ void jbd2_journal_commit_transaction(journal_t *journal) if (bufs == journal->j_wbufsize || commit_transaction->t_buffers == NULL || - space_left < tag_bytes + 16) { + space_left < tag_bytes + 16 + csum_size) { - jbd_debug(4, "JBD: Submit %d IOs\n", bufs); + jbd_debug(4, "JBD2: 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(JBD2_FLAG_LAST_TAG); + tag->t_flags |= cpu_to_be16(JBD2_FLAG_LAST_TAG); + jbd2_descr_block_csum_set(journal, descriptor); start_journal_io: for (i = 0; i < bufs; i++) { struct buffer_head *bh = wbuf[i]; + /* + * Compute checksum. + */ + if (JBD2_HAS_COMPAT_FEATURE(journal, + JBD2_FEATURE_COMPAT_CHECKSUM)) { + crc32_sum = + jbd2_checksum_data(crc32_sum, bh); + } + lock_buffer(bh); clear_buffer_dirty(bh); set_buffer_uptodate(bh); bh->b_end_io = journal_end_buffer_io_sync; - submit_bh(WRITE, bh); + submit_bh(WRITE_SYNC, bh); } cond_resched(); + stats.run.rs_blocks_logged += bufs; /* Force a new descriptor to be generated next time round the loop. */ @@ -655,60 +751,101 @@ start_journal_io: } } + err = journal_finish_inode_data_buffers(journal, commit_transaction); + if (err) { + printk(KERN_WARNING + "JBD2: Detected IO errors while flushing file data " + "on %s\n", journal->j_devname); + if (journal->j_flags & JBD2_ABORT_ON_SYNCDATA_ERR) + jbd2_journal_abort(journal, err); + 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 + * the commit record + */ + if (commit_transaction->t_need_data_flush && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS, 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, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + } + + blk_finish_plug(&plug); + /* 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. + the io_bufs list. 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"); + jbd_debug(3, "JBD2: commit phase 3\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; + while (!list_empty(&io_bufs)) { + struct buffer_head *bh = list_entry(io_bufs.prev, + struct buffer_head, + b_assoc_buffers); - 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; + wait_on_buffer(bh); + cond_resched(); if (unlikely(!buffer_uptodate(bh))) err = -EIO; - - clear_buffer_jwrite(bh); - - JBUFFER_TRACE(jh, "ph4: unfile after journal write"); - jbd2_journal_unfile_buffer(journal, jh); + jbd2_unfile_log_bh(bh); /* - * ->t_iobuf_list should contain only dummy buffer_heads - * which were created by jbd2_journal_write_metadata_buffer(). + * The list contains temporary buffer heads created by + * jbd2_journal_write_metadata_buffer(). */ BUFFER_TRACE(bh, "dumping temporary bh"); - jbd2_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 */ + /* We also have to refile the corresponding shadowed buffer */ jh = commit_transaction->t_shadow_list->b_tprev; bh = jh2bh(jh); - clear_bit(BH_JWrite, &bh->b_state); + clear_buffer_jwrite(bh); J_ASSERT_BH(bh, buffer_jbddirty(bh)); + J_ASSERT_BH(bh, !buffer_shadow(bh)); /* The metadata is now released for reuse, but we need to remember it against this transaction so that when @@ -716,63 +853,78 @@ wait_for_iobuf: required. */ JBUFFER_TRACE(jh, "file as BJ_Forget"); jbd2_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"); + jbd_debug(3, "JBD2: commit phase 4\n"); /* Here we wait for the revoke record and descriptor record buffers */ - wait_for_ctlbuf: - while (commit_transaction->t_log_list != NULL) { + while (!list_empty(&log_bufs)) { 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; + bh = list_entry(log_bufs.prev, struct buffer_head, b_assoc_buffers); + wait_on_buffer(bh); + cond_resched(); if (unlikely(!buffer_uptodate(bh))) err = -EIO; BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile"); clear_buffer_jwrite(bh); - jbd2_journal_unfile_buffer(journal, jh); - jbd2_journal_put_journal_head(jh); + jbd2_unfile_log_bh(bh); __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) + jbd2_journal_abort(journal, err); + + jbd_debug(3, "JBD2: commit phase 5\n"); + write_lock(&journal->j_state_lock); + J_ASSERT(commit_transaction->t_state == T_COMMIT_DFLUSH); + commit_transaction->t_state = T_COMMIT_JFLUSH; + write_unlock(&journal->j_state_lock); + + if (!JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { + err = journal_submit_commit_record(journal, commit_transaction, + &cbh, crc32_sum); + if (err) + __jbd2_journal_abort_hard(journal); + } + if (cbh) + err = journal_wait_on_commit_record(journal, cbh); + if (JBD2_HAS_INCOMPAT_FEATURE(journal, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT) && + journal->j_flags & JBD2_BARRIER) { + blkdev_issue_flush(journal->j_dev, GFP_NOFS, NULL); + } if (err) - __jbd2_journal_abort_hard(journal); + 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 before. */ - jbd_debug(3, "JBD: commit phase 7\n"); + jbd_debug(3, "JBD2: commit phase 6\n"); - J_ASSERT(commit_transaction->t_sync_datalist == NULL); + J_ASSERT(list_empty(&commit_transaction->t_inode_list)); 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: /* @@ -783,13 +935,18 @@ restart_loop: while (commit_transaction->t_forget) { transaction_t *cp_transaction; struct buffer_head *bh; + int try_to_free = 0; jh = commit_transaction->t_forget; spin_unlock(&journal->j_list_lock); bh = jh2bh(jh); + /* + * Get a reference so that bh cannot be freed before we are + * done with it. + */ + get_bh(bh); 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 @@ -800,23 +957,29 @@ restart_loop: * data. * * Otherwise, we can just throw away the frozen data now. + * + * We also know that the frozen data has already fired + * its triggers if they exist, so we can clear that too. */ if (jh->b_committed_data) { - jbd2_slab_free(jh->b_committed_data, bh->b_size); + jbd2_free(jh->b_committed_data, bh->b_size); jh->b_committed_data = NULL; if (jh->b_frozen_data) { jh->b_committed_data = jh->b_frozen_data; jh->b_frozen_data = NULL; + jh->b_frozen_triggers = NULL; } } else if (jh->b_frozen_data) { - jbd2_slab_free(jh->b_frozen_data, bh->b_size); + jbd2_free(jh->b_frozen_data, bh->b_size); jh->b_frozen_data = NULL; + jh->b_frozen_triggers = NULL; } spin_lock(&journal->j_list_lock); cp_transaction = jh->b_cp_transaction; if (cp_transaction) { JBUFFER_TRACE(jh, "remove from old cp transaction"); + cp_transaction->t_chp_stats.cs_dropped++; __jbd2_journal_remove_checkpoint(jh); } @@ -826,54 +989,73 @@ restart_loop: * 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. */ + /* + * A buffer which has been freed while still being journaled by + * a previous transaction. + */ if (buffer_freed(bh)) { - clear_buffer_freed(bh); - clear_buffer_jbddirty(bh); + /* + * If the running transaction is the one containing + * "add to orphan" operation (b_next_transaction != + * NULL), we have to wait for that transaction to + * commit before we can really get rid of the buffer. + * So just clear b_modified to not confuse transaction + * credit accounting and refile the buffer to + * BJ_Forget of the running transaction. If the just + * committed transaction contains "add to orphan" + * operation, we can completely invalidate the buffer + * now. We are rather through in that since the + * buffer may be still accessible when blocksize < + * pagesize and it is attached to the last partial + * page. + */ + jh->b_modified = 0; + if (!jh->b_next_transaction) { + clear_buffer_freed(bh); + clear_buffer_jbddirty(bh); + clear_buffer_mapped(bh); + clear_buffer_new(bh); + clear_buffer_req(bh); + bh->b_bdev = NULL; + } } if (buffer_jbddirty(bh)) { JBUFFER_TRACE(jh, "add to new checkpointing trans"); __jbd2_journal_insert_checkpoint(jh, commit_transaction); - JBUFFER_TRACE(jh, "refile for checkpoint writeback"); - __jbd2_journal_refile_buffer(jh); - jbd_unlock_bh_state(bh); + if (is_journal_aborted(journal)) + clear_buffer_jbddirty(bh); } else { J_ASSERT_BH(bh, !buffer_dirty(bh)); - /* The buffer on BJ_Forget list and not jbddirty means + /* + * The buffer on BJ_Forget list and not jbddirty means * it has been freed by this transaction and hence it * could not have been reallocated until this * transaction has committed. *BUT* it could be * reallocated once we have written all the data to * disk and before we process the buffer on BJ_Forget - * list. */ - JBUFFER_TRACE(jh, "refile or unfile freed buffer"); - __jbd2_journal_refile_buffer(jh); - if (!jh->b_transaction) { - jbd_unlock_bh_state(bh); - /* needs a brelse */ - jbd2_journal_remove_journal_head(bh); - release_buffer_page(bh); - } else - jbd_unlock_bh_state(bh); + * list. + */ + if (!jh->b_next_transaction) + try_to_free = 1; } + JBUFFER_TRACE(jh, "refile or unfile buffer"); + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + if (try_to_free) + release_buffer_page(bh); /* Drops bh reference */ + else + __brelse(bh); cond_resched_lock(&journal->j_list_lock); } spin_unlock(&journal->j_list_lock); /* - * This is a bit sleazy. We borrow j_list_lock to protect - * journal->j_committing_transaction in __jbd2_journal_remove_checkpoint. - * Really, __jbd2_journal_remove_checkpoint should be using j_state_lock but - * it's a bit hassle to hold that across __jbd2_journal_remove_checkpoint + * This is a bit sleazy. We use j_list_lock to protect transition + * of a transaction into T_FINISHED state and calling + * __jbd2_journal_drop_transaction(). Otherwise we could race with + * other checkpointing code processing the transaction... */ - spin_lock(&journal->j_state_lock); + write_lock(&journal->j_state_lock); spin_lock(&journal->j_list_lock); /* * Now recheck if some buffers did not get attached to the transaction @@ -881,44 +1063,100 @@ restart_loop: */ if (commit_transaction->t_forget) { spin_unlock(&journal->j_list_lock); - spin_unlock(&journal->j_state_lock); + write_unlock(&journal->j_state_lock); goto restart_loop; } /* Done with this transaction! */ - jbd_debug(3, "JBD: commit phase 8\n"); + jbd_debug(3, "JBD2: commit phase 7\n"); - J_ASSERT(commit_transaction->t_state == T_COMMIT); + J_ASSERT(commit_transaction->t_state == T_COMMIT_JFLUSH); - commit_transaction->t_state = T_FINISHED; + commit_transaction->t_start = jiffies; + stats.run.rs_logging = jbd2_time_diff(stats.run.rs_logging, + commit_transaction->t_start); + + /* + * File the transaction statistics + */ + stats.ts_tid = commit_transaction->t_tid; + stats.run.rs_handle_count = + atomic_read(&commit_transaction->t_handle_count); + trace_jbd2_run_stats(journal->j_fs_dev->bd_dev, + commit_transaction->t_tid, &stats.run); + + /* + * Calculate overall stats + */ + spin_lock(&journal->j_history_lock); + journal->j_stats.ts_tid++; + if (commit_transaction->t_requested) + journal->j_stats.ts_requested++; + journal->j_stats.run.rs_wait += stats.run.rs_wait; + journal->j_stats.run.rs_request_delay += stats.run.rs_request_delay; + journal->j_stats.run.rs_running += stats.run.rs_running; + journal->j_stats.run.rs_locked += stats.run.rs_locked; + journal->j_stats.run.rs_flushing += stats.run.rs_flushing; + journal->j_stats.run.rs_logging += stats.run.rs_logging; + journal->j_stats.run.rs_handle_count += stats.run.rs_handle_count; + journal->j_stats.run.rs_blocks += stats.run.rs_blocks; + journal->j_stats.run.rs_blocks_logged += stats.run.rs_blocks_logged; + spin_unlock(&journal->j_history_lock); + + commit_transaction->t_state = T_COMMIT_CALLBACK; 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); + commit_time = ktime_to_ns(ktime_sub(ktime_get(), start_time)); - if (commit_transaction->t_checkpoint_list == NULL) { - __jbd2_journal_drop_transaction(journal, commit_transaction); + /* + * weight the commit time higher than the average time so we don't + * react too strongly to vast changes in the commit time + */ + if (likely(journal->j_average_commit_time)) + journal->j_average_commit_time = (commit_time + + journal->j_average_commit_time*3) / 4; + else + journal->j_average_commit_time = commit_time; + + write_unlock(&journal->j_state_lock); + + 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 { - 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->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", + /* Drop all spin_locks because commit_callback may be block. + * __journal_remove_checkpoint() can not destroy transaction + * under us because it is not marked as T_FINISHED yet */ + if (journal->j_commit_callback) + journal->j_commit_callback(journal, commit_transaction); + + trace_jbd2_end_commit(journal, commit_transaction); + jbd_debug(1, "JBD2: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); + write_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + commit_transaction->t_state = T_FINISHED; + /* Recheck checkpoint lists after j_list_lock was dropped */ + if (commit_transaction->t_checkpoint_list == NULL && + commit_transaction->t_checkpoint_io_list == NULL) { + __jbd2_journal_drop_transaction(journal, commit_transaction); + jbd2_journal_free_transaction(commit_transaction); + } + spin_unlock(&journal->j_list_lock); + write_unlock(&journal->j_state_lock); wake_up(&journal->j_wait_done_commit); } |