From e9e34f4e8f42177c66754fec1edfd35e70c18f99 Mon Sep 17 00:00:00 2001 From: Hidehiro Kawai Date: Thu, 31 Jul 2008 22:26:04 -0400 Subject: jbd2: don't abort if flushing file data failed In ordered mode, the current jbd2 aborts the journal if a file data buffer has an error. But this behavior is unintended, and we found that it has been adopted accidentally. This patch undoes it and just calls printk() instead of aborting the journal. Unlike a similar patch for ext3/jbd, file data buffers are written via generic_writepages(). But we also need to set AS_EIO into their mappings because wait_on_page_writeback_range() clears AS_EIO before a user process sees it. Signed-off-by: Hidehiro Kawai Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index f8b3be873226..adf0395f318e 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -262,8 +262,18 @@ static int journal_finish_inode_data_buffers(journal_t *journal, jinode->i_flags |= JI_COMMIT_RUNNING; spin_unlock(&journal->j_list_lock); err = filemap_fdatawait(jinode->i_vfs_inode->i_mapping); - if (!ret) - ret = err; + if (err) { + /* + * Because AS_EIO is cleared by + * wait_on_page_writeback_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); jinode->i_flags &= ~JI_COMMIT_RUNNING; wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); @@ -670,8 +680,14 @@ start_journal_io: * commit block, which happens below in such setting. */ err = journal_finish_inode_data_buffers(journal, commit_transaction); - if (err) - jbd2_journal_abort(journal, err); + if (err) { + char b[BDEVNAME_SIZE]; + + printk(KERN_WARNING + "JBD2: Detected IO errors while flushing file data " + "on %s\n", bdevname(journal->j_fs_dev, b)); + err = 0; + } /* 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 -- cgit v1.2.3 From 529ae9aaa08378cfe2a4350bded76f32cc8ff0ce Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 2 Aug 2008 12:01:03 +0200 Subject: mm: rename page trylock Converting page lock to new locking bitops requires a change of page flag operation naming, so we might as well convert it to something nicer (!TestSetPageLocked_Lock => trylock_page, SetPageLocked => set_page_locked). This also facilitates lockdeping of page lock. Signed-off-by: Nick Piggin Acked-by: KOSAKI Motohiro Acked-by: Peter Zijlstra Acked-by: Andrew Morton Acked-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- drivers/scsi/sg.c | 2 +- fs/afs/write.c | 2 +- fs/cifs/file.c | 2 +- fs/jbd/commit.c | 4 +-- fs/jbd2/commit.c | 2 +- fs/reiserfs/journal.c | 2 +- fs/splice.c | 2 +- fs/xfs/linux-2.6/xfs_aops.c | 4 +-- include/linux/page-flags.h | 2 +- include/linux/pagemap.h | 67 +++++++++++++++++++++++++++------------------ mm/filemap.c | 12 ++++---- mm/memory.c | 2 +- mm/migrate.c | 4 +-- mm/rmap.c | 2 +- mm/shmem.c | 4 +-- mm/swap.c | 2 +- mm/swap_state.c | 8 +++--- mm/swapfile.c | 2 +- mm/truncate.c | 4 +-- mm/vmscan.c | 4 +-- 20 files changed, 74 insertions(+), 59 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c index d3b8ebb83776..3d36270a8b4d 100644 --- a/drivers/scsi/sg.c +++ b/drivers/scsi/sg.c @@ -1747,7 +1747,7 @@ st_map_user_pages(struct scatterlist *sgl, const unsigned int max_pages, */ flush_dcache_page(pages[i]); /* ?? Is locking needed? I don't think so */ - /* if (TestSetPageLocked(pages[i])) + /* if (!trylock_page(pages[i])) goto out_unlock; */ } diff --git a/fs/afs/write.c b/fs/afs/write.c index 9a849ad3c489..065b4e10681a 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -404,7 +404,7 @@ static int afs_write_back_from_locked_page(struct afs_writeback *wb, page = pages[loop]; if (page->index > wb->last) break; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) break; if (!PageDirty(page) || page_private(page) != (unsigned long) wb) { diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 0aac824371a5..e692c42f24b5 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1280,7 +1280,7 @@ retry: if (first < 0) lock_page(page); - else if (TestSetPageLocked(page)) + else if (!trylock_page(page)) break; if (unlikely(page->mapping != mapping)) { diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 2eccbfaa1d48..81a9ad7177ca 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -63,7 +63,7 @@ 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); @@ -446,7 +446,7 @@ void journal_commit_transaction(journal_t *journal) spin_lock(&journal->j_list_lock); } if (unlikely(!buffer_uptodate(bh))) { - if (TestSetPageLocked(bh->b_page)) { + if (!trylock_page(bh->b_page)) { spin_unlock(&journal->j_list_lock); lock_page(bh->b_page); spin_lock(&journal->j_list_lock); diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index adf0395f318e..f2ad061e95ec 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -67,7 +67,7 @@ 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); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index c8f60ee183b5..ce2208b27118 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -627,7 +627,7 @@ static int journal_list_still_alive(struct super_block *s, static void release_buffer_page(struct buffer_head *bh) { struct page *page = bh->b_page; - if (!page->mapping && !TestSetPageLocked(page)) { + if (!page->mapping && trylock_page(page)) { page_cache_get(page); put_bh(bh); if (!page->mapping) diff --git a/fs/splice.c b/fs/splice.c index b30311ba8af6..1bbc6f4bb09c 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -371,7 +371,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * for an in-flight io page */ if (flags & SPLICE_F_NONBLOCK) { - if (TestSetPageLocked(page)) { + if (!trylock_page(page)) { error = -EAGAIN; break; } diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index 0b211cba1909..fa73179233ad 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -675,7 +675,7 @@ xfs_probe_cluster( } else pg_offset = PAGE_CACHE_SIZE; - if (page->index == tindex && !TestSetPageLocked(page)) { + if (page->index == tindex && trylock_page(page)) { pg_len = xfs_probe_page(page, pg_offset, mapped); unlock_page(page); } @@ -759,7 +759,7 @@ xfs_convert_page( if (page->index != tindex) goto fail; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto fail; if (PageWriteback(page)) goto fail_unlock_page; diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 25aaccdb2f26..c74d3e875314 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -163,7 +163,7 @@ static inline int Page##uname(struct page *page) \ struct page; /* forward declaration */ -PAGEFLAG(Locked, locked) TESTSCFLAG(Locked, locked) +TESTPAGEFLAG(Locked, locked) PAGEFLAG(Error, error) PAGEFLAG(Referenced, referenced) TESTCLEARFLAG(Referenced, referenced) PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty) diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index 69ed3cb1197a..5da31c12101c 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -250,29 +250,6 @@ static inline struct page *read_mapping_page(struct address_space *mapping, return read_cache_page(mapping, index, filler, data); } -int add_to_page_cache_locked(struct page *page, struct address_space *mapping, - pgoff_t index, gfp_t gfp_mask); -int add_to_page_cache_lru(struct page *page, struct address_space *mapping, - pgoff_t index, gfp_t gfp_mask); -extern void remove_from_page_cache(struct page *page); -extern void __remove_from_page_cache(struct page *page); - -/* - * Like add_to_page_cache_locked, but used to add newly allocated pages: - * the page is new, so we can just run SetPageLocked() against it. - */ -static inline int add_to_page_cache(struct page *page, - struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask) -{ - int error; - - SetPageLocked(page); - error = add_to_page_cache_locked(page, mapping, offset, gfp_mask); - if (unlikely(error)) - ClearPageLocked(page); - return error; -} - /* * Return byte-offset into filesystem object for page. */ @@ -294,13 +271,28 @@ extern int __lock_page_killable(struct page *page); extern void __lock_page_nosync(struct page *page); extern void unlock_page(struct page *page); +static inline void set_page_locked(struct page *page) +{ + set_bit(PG_locked, &page->flags); +} + +static inline void clear_page_locked(struct page *page) +{ + clear_bit(PG_locked, &page->flags); +} + +static inline int trylock_page(struct page *page) +{ + return !test_and_set_bit(PG_locked, &page->flags); +} + /* * lock_page may only be called if we have the page's inode pinned. */ static inline void lock_page(struct page *page) { might_sleep(); - if (TestSetPageLocked(page)) + if (!trylock_page(page)) __lock_page(page); } @@ -312,7 +304,7 @@ static inline void lock_page(struct page *page) static inline int lock_page_killable(struct page *page) { might_sleep(); - if (TestSetPageLocked(page)) + if (!trylock_page(page)) return __lock_page_killable(page); return 0; } @@ -324,7 +316,7 @@ static inline int lock_page_killable(struct page *page) static inline void lock_page_nosync(struct page *page) { might_sleep(); - if (TestSetPageLocked(page)) + if (!trylock_page(page)) __lock_page_nosync(page); } @@ -409,4 +401,27 @@ static inline int fault_in_pages_readable(const char __user *uaddr, int size) return ret; } +int add_to_page_cache_locked(struct page *page, struct address_space *mapping, + pgoff_t index, gfp_t gfp_mask); +int add_to_page_cache_lru(struct page *page, struct address_space *mapping, + pgoff_t index, gfp_t gfp_mask); +extern void remove_from_page_cache(struct page *page); +extern void __remove_from_page_cache(struct page *page); + +/* + * Like add_to_page_cache_locked, but used to add newly allocated pages: + * the page is new, so we can just run set_page_locked() against it. + */ +static inline int add_to_page_cache(struct page *page, + struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask) +{ + int error; + + set_page_locked(page); + error = add_to_page_cache_locked(page, mapping, offset, gfp_mask); + if (unlikely(error)) + clear_page_locked(page); + return error; +} + #endif /* _LINUX_PAGEMAP_H */ diff --git a/mm/filemap.c b/mm/filemap.c index d97d1ad55473..54e968650855 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -558,14 +558,14 @@ EXPORT_SYMBOL(wait_on_page_bit); * But that's OK - sleepers in wait_on_page_writeback() just go back to sleep. * * The first mb is necessary to safely close the critical section opened by the - * TestSetPageLocked(), the second mb is necessary to enforce ordering between - * the clear_bit and the read of the waitqueue (to avoid SMP races with a - * parallel wait_on_page_locked()). + * test_and_set_bit() to lock the page; the second mb is necessary to enforce + * ordering between the clear_bit and the read of the waitqueue (to avoid SMP + * races with a parallel wait_on_page_locked()). */ void unlock_page(struct page *page) { smp_mb__before_clear_bit(); - if (!TestClearPageLocked(page)) + if (!test_and_clear_bit(PG_locked, &page->flags)) BUG(); smp_mb__after_clear_bit(); wake_up_page(page, PG_locked); @@ -931,7 +931,7 @@ grab_cache_page_nowait(struct address_space *mapping, pgoff_t index) struct page *page = find_get_page(mapping, index); if (page) { - if (!TestSetPageLocked(page)) + if (trylock_page(page)) return page; page_cache_release(page); return NULL; @@ -1027,7 +1027,7 @@ find_page: if (inode->i_blkbits == PAGE_CACHE_SHIFT || !mapping->a_ops->is_partially_uptodate) goto page_not_up_to_date; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto page_not_up_to_date; if (!mapping->a_ops->is_partially_uptodate(page, desc, offset)) diff --git a/mm/memory.c b/mm/memory.c index a472bcd4b061..1002f473f497 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1789,7 +1789,7 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, * not dirty accountable. */ if (PageAnon(old_page)) { - if (!TestSetPageLocked(old_page)) { + if (trylock_page(old_page)) { reuse = can_share_swap_page(old_page); unlock_page(old_page); } diff --git a/mm/migrate.c b/mm/migrate.c index 153572fb60b8..2a80136b23bb 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -605,7 +605,7 @@ static int move_to_new_page(struct page *newpage, struct page *page) * establishing additional references. We are the only one * holding a reference to the new page at this point. */ - if (TestSetPageLocked(newpage)) + if (!trylock_page(newpage)) BUG(); /* Prepare mapping for the new page.*/ @@ -667,7 +667,7 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private, BUG_ON(charge); rc = -EAGAIN; - if (TestSetPageLocked(page)) { + if (!trylock_page(page)) { if (!force) goto move_newpage; lock_page(page); diff --git a/mm/rmap.c b/mm/rmap.c index 94a5246a3f98..1ea4e6fcee77 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -422,7 +422,7 @@ int page_referenced(struct page *page, int is_locked, referenced += page_referenced_anon(page, mem_cont); else if (is_locked) referenced += page_referenced_file(page, mem_cont); - else if (TestSetPageLocked(page)) + else if (!trylock_page(page)) referenced++; else { if (page->mapping) diff --git a/mm/shmem.c b/mm/shmem.c index c1e5a3b4f758..04fb4f1ab88e 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1265,7 +1265,7 @@ repeat: } /* We have to do this with page locked to prevent races */ - if (TestSetPageLocked(swappage)) { + if (!trylock_page(swappage)) { shmem_swp_unmap(entry); spin_unlock(&info->lock); wait_on_page_locked(swappage); @@ -1329,7 +1329,7 @@ repeat: shmem_swp_unmap(entry); filepage = find_get_page(mapping, idx); if (filepage && - (!PageUptodate(filepage) || TestSetPageLocked(filepage))) { + (!PageUptodate(filepage) || !trylock_page(filepage))) { spin_unlock(&info->lock); wait_on_page_locked(filepage); page_cache_release(filepage); diff --git a/mm/swap.c b/mm/swap.c index 7417a2adbe50..9e0cb3118079 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -444,7 +444,7 @@ void pagevec_strip(struct pagevec *pvec) for (i = 0; i < pagevec_count(pvec); i++) { struct page *page = pvec->pages[i]; - if (PagePrivate(page) && !TestSetPageLocked(page)) { + if (PagePrivate(page) && trylock_page(page)) { if (PagePrivate(page)) try_to_release_page(page, 0); unlock_page(page); diff --git a/mm/swap_state.c b/mm/swap_state.c index b8035b055129..167cf2dc8a03 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -201,7 +201,7 @@ void delete_from_swap_cache(struct page *page) */ static inline void free_swap_cache(struct page *page) { - if (PageSwapCache(page) && !TestSetPageLocked(page)) { + if (PageSwapCache(page) && trylock_page(page)) { remove_exclusive_swap_page(page); unlock_page(page); } @@ -302,9 +302,9 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, * re-using the just freed swap entry for an existing page. * May fail (-ENOMEM) if radix-tree node allocation failed. */ - SetPageLocked(new_page); + set_page_locked(new_page); err = add_to_swap_cache(new_page, entry, gfp_mask & GFP_KERNEL); - if (!err) { + if (likely(!err)) { /* * Initiate read into locked page and return. */ @@ -312,7 +312,7 @@ struct page *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask, swap_readpage(NULL, new_page); return new_page; } - ClearPageLocked(new_page); + clear_page_locked(new_page); swap_free(entry); } while (err != -ENOMEM); diff --git a/mm/swapfile.c b/mm/swapfile.c index bb7f79641f9e..1e330f2998fa 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -403,7 +403,7 @@ void free_swap_and_cache(swp_entry_t entry) if (p) { if (swap_entry_free(p, swp_offset(entry)) == 1) { page = find_get_page(&swapper_space, entry.val); - if (page && unlikely(TestSetPageLocked(page))) { + if (page && unlikely(!trylock_page(page))) { page_cache_release(page); page = NULL; } diff --git a/mm/truncate.c b/mm/truncate.c index 894e9a70699f..250505091d37 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -187,7 +187,7 @@ void truncate_inode_pages_range(struct address_space *mapping, if (page_index > next) next = page_index; next++; - if (TestSetPageLocked(page)) + if (!trylock_page(page)) continue; if (PageWriteback(page)) { unlock_page(page); @@ -280,7 +280,7 @@ unsigned long __invalidate_mapping_pages(struct address_space *mapping, pgoff_t index; int lock_failed; - lock_failed = TestSetPageLocked(page); + lock_failed = !trylock_page(page); /* * We really shouldn't be looking at the ->index of an diff --git a/mm/vmscan.c b/mm/vmscan.c index 75be453628bf..1ff1a58e7c10 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -496,7 +496,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, page = lru_to_page(page_list); list_del(&page->lru); - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto keep; VM_BUG_ON(PageActive(page)); @@ -582,7 +582,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, * A synchronous write - probably a ramdisk. Go * ahead and try to reclaim the page. */ - if (TestSetPageLocked(page)) + if (!trylock_page(page)) goto keep; if (PageDirty(page) || PageWriteback(page)) goto keep_locked; -- cgit v1.2.3 From 05496769e5da83ce22ed97345afd9c7b71d6bd24 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 16 Sep 2008 14:36:17 -0400 Subject: jbd2: clean up how the journal device name is printed Calculate the journal device name once and stash it away in the journal_s structure. This avoids needing to call bdevname() everywhere and reduces stack usage by not needing to allocate an on-stack buffer. In addition, we eliminate the '/' that can appear in device names (e.g. "cciss/c0d0p9" --- see kernel bugzilla #11321) that can cause problems when creating proc directory names, and include the inode number to support ocfs2 which creates multiple journals with different inode numbers. Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 12 +++--------- fs/jbd2/commit.c | 11 +++-------- fs/jbd2/journal.c | 48 ++++++++++++++++-------------------------------- include/linux/jbd2.h | 3 ++- 4 files changed, 24 insertions(+), 50 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f58cc0309dc9..64e1c21eb5d2 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1476,15 +1476,9 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt); - if (EXT4_SB(sb)->s_journal->j_inode == NULL) { - char b[BDEVNAME_SIZE]; - - printk(KERN_INFO "EXT4 FS on %s, external journal on %s\n", - sb->s_id, bdevname(EXT4_SB(sb)->s_journal->j_dev, b)); - } else { - printk(KERN_INFO "EXT4 FS on %s, internal journal\n", - sb->s_id); - } + printk(KERN_INFO "EXT4 FS on %s, %s journal on %s\n", + sb->s_id, EXT4_SB(sb)->s_journal->j_inode ? "internal" : + "external", EXT4_SB(sb)->s_journal->j_devname); return res; } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index f2ad061e95ec..b091e5378fe0 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -147,12 +147,9 @@ static int journal_submit_commit_record(journal_t *journal, * to remember if we sent a barrier request */ if (ret == -EOPNOTSUPP && barrier_done) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING - "JBD: barrier-based sync failed on %s - " - "disabling barriers\n", - bdevname(journal->j_dev, b)); + "JBD: barrier-based sync failed on %s - " + "disabling barriers\n", journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags &= ~JBD2_BARRIER; spin_unlock(&journal->j_state_lock); @@ -681,11 +678,9 @@ start_journal_io: */ err = journal_finish_inode_data_buffers(journal, commit_transaction); if (err) { - char b[BDEVNAME_SIZE]; - printk(KERN_WARNING "JBD2: Detected IO errors while flushing file data " - "on %s\n", bdevname(journal->j_fs_dev, b)); + "on %s\n", journal->j_devname); err = 0; } diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 8207a01c4edb..81186a29742e 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -597,13 +597,9 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, if (ret) *retp = ret; else { - char b[BDEVNAME_SIZE]; - printk(KERN_ALERT "%s: journal block not found " "at offset %lu on %s\n", - __func__, - blocknr, - bdevname(journal->j_dev, b)); + __func__, blocknr, journal->j_devname); err = -EIO; __journal_abort_soft(journal, err); } @@ -901,10 +897,7 @@ static struct proc_dir_entry *proc_jbd2_stats; static void jbd2_stats_proc_init(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); - journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats); + journal->j_proc_entry = proc_mkdir(journal->j_devname, proc_jbd2_stats); if (journal->j_proc_entry) { proc_create_data("history", S_IRUGO, journal->j_proc_entry, &jbd2_seq_history_fops, journal); @@ -915,12 +908,9 @@ static void jbd2_stats_proc_init(journal_t *journal) static void jbd2_stats_proc_exit(journal_t *journal) { - char name[BDEVNAME_SIZE]; - - bdevname(journal->j_dev, name); remove_proc_entry("info", journal->j_proc_entry); remove_proc_entry("history", journal->j_proc_entry); - remove_proc_entry(name, proc_jbd2_stats); + remove_proc_entry(journal->j_devname, proc_jbd2_stats); } static void journal_init_stats(journal_t *journal) @@ -1018,6 +1008,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, { journal_t *journal = journal_init_common(); struct buffer_head *bh; + char *p; int n; if (!journal) @@ -1039,6 +1030,10 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, journal->j_fs_dev = fs_dev; journal->j_blk_offset = start; journal->j_maxlen = len; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; jbd2_stats_proc_init(journal); bh = __getblk(journal->j_dev, start, journal->j_blocksize); @@ -1061,6 +1056,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) { struct buffer_head *bh; journal_t *journal = journal_init_common(); + char *p; int err; int n; unsigned long long blocknr; @@ -1070,6 +1066,12 @@ journal_t * jbd2_journal_init_inode (struct inode *inode) journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev; journal->j_inode = inode; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; + p = journal->j_devname + strlen(journal->j_devname); + sprintf(p, ":%lu", journal->j_inode->i_ino); jbd_debug(1, "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n", journal, inode->i_sb->s_id, inode->i_ino, @@ -1760,23 +1762,6 @@ int jbd2_journal_wipe(journal_t *journal, int write) return err; } -/* - * journal_dev_name: format a character string to describe on what - * device this journal is present. - */ - -static const char *journal_dev_name(journal_t *journal, char *buffer) -{ - struct block_device *bdev; - - if (journal->j_inode) - bdev = journal->j_inode->i_sb->s_bdev; - else - bdev = journal->j_dev; - - return bdevname(bdev, buffer); -} - /* * Journal abort has very specific semantics, which we describe * for journal abort. @@ -1793,13 +1778,12 @@ static const char *journal_dev_name(journal_t *journal, char *buffer) void __jbd2_journal_abort_hard(journal_t *journal) { transaction_t *transaction; - char b[BDEVNAME_SIZE]; if (journal->j_flags & JBD2_ABORT) return; printk(KERN_ERR "Aborting journal on device %s.\n", - journal_dev_name(journal, b)); + journal->j_devname); spin_lock(&journal->j_state_lock); journal->j_flags |= JBD2_ABORT; diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 3dd209007098..66c3499478b5 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -850,7 +850,8 @@ struct journal_s */ struct block_device *j_dev; int j_blocksize; - unsigned long long j_blk_offset; + unsigned long long j_blk_offset; + char j_devname[BDEVNAME_SIZE+24]; /* * Device which holds the client fs. For internal journal this will be -- cgit v1.2.3 From ede86cc473defab74d778aeac14b19f43129d4d1 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sun, 5 Oct 2008 20:50:06 -0400 Subject: ext4: Add debugging markers that can be used by systemtap This debugging markers are designed to debug problems such as the random filesystem latency problems reported by Arjan. Signed-off-by: "Theodore Ts'o" --- fs/ext4/fsync.c | 5 +++++ fs/ext4/super.c | 2 ++ fs/jbd2/checkpoint.c | 3 +++ fs/jbd2/commit.c | 6 ++++++ 4 files changed, 16 insertions(+) (limited to 'fs/jbd2/commit.c') diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index c37d1e86f51a..5afe4370840b 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -28,6 +28,7 @@ #include #include #include +#include #include "ext4.h" #include "ext4_jbd2.h" @@ -51,6 +52,10 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) J_ASSERT(ext4_journal_current_handle() == NULL); + trace_mark(ext4_sync_file, "dev %s datasync %d ino %ld parent %ld", + inode->i_sb->s_id, datasync, inode->i_ino, + dentry->d_parent->d_inode->i_ino); + /* * data=writeback: * The caller's filemap_fdatawrite()/wait will sync the data. diff --git a/fs/ext4/super.c b/fs/ext4/super.c index dfcd41fafb9f..9c0214689de0 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -35,6 +35,7 @@ #include #include #include +#include #include #include #include @@ -2951,6 +2952,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait) { tid_t target; + trace_mark(ext4_sync_fs, "dev %s wait %d", sb->s_id, wait); sb->s_dirt = 0; if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { if (wait) diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index af4651bf3570..42895d369458 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -328,6 +329,8 @@ int jbd2_log_do_checkpoint(journal_t *journal) * journal straight away. */ result = jbd2_cleanup_journal_tail(journal); + trace_mark(jbd2_checkpoint, "dev %s need_checkpoint %d", + journal->j_devname, result); jbd_debug(1, "cleanup_journal_tail returned %d\n", result); if (result <= 0) return result; diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index b091e5378fe0..e91f051a9859 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -368,6 +369,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) commit_transaction = journal->j_running_transaction; J_ASSERT(commit_transaction->t_state == T_RUNNING); + trace_mark(jbd2_start_commit, "dev %s transaction %d", + journal->j_devname, commit_transaction->t_tid); jbd_debug(1, "JBD: starting commit of transaction %d\n", commit_transaction->t_tid); @@ -985,6 +988,9 @@ restart_loop: } spin_unlock(&journal->j_list_lock); + trace_mark(jbd2_end_commit, "dev %s transaction %d head %d", + journal->j_devname, commit_transaction->t_tid, + journal->j_tail_sequence); jbd_debug(1, "JBD: commit %d complete, head %d\n", journal->j_commit_sequence, journal->j_tail_sequence); -- cgit v1.2.3 From 45a90bfd90c1215bf824c0f705b409723f52361b Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 6 Oct 2008 12:04:02 -0400 Subject: jbd2: Fix buffer head leak when writing the commit block Also make sure the buffer heads are marked clean before submitting bh for writing. The previous code was marking the buffer head dirty, which would have forced an unneeded write (and seek) to the journal for no good reason. Signed-off-by: "Theodore Ts'o" --- fs/jbd2/commit.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index e91f051a9859..0d3814a35ed1 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -127,8 +127,7 @@ static int journal_submit_commit_record(journal_t *journal, JBUFFER_TRACE(descriptor, "submit commit block"); lock_buffer(bh); - get_bh(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); set_buffer_uptodate(bh); bh->b_end_io = journal_end_buffer_io_sync; @@ -158,7 +157,7 @@ static int journal_submit_commit_record(journal_t *journal, /* And try again, without the barrier */ lock_buffer(bh); set_buffer_uptodate(bh); - set_buffer_dirty(bh); + clear_buffer_dirty(bh); ret = submit_bh(WRITE, bh); } *cbh = bh; -- cgit v1.2.3 From 77e841de8abac4755cc83ca224fdf71418d65380 Mon Sep 17 00:00:00 2001 From: Hidehiro Kawai Date: Sun, 12 Oct 2008 16:39:16 -0400 Subject: jbd2: abort when failed to log metadata buffers If we failed to write metadata buffers to the journal space and succeeded to write the commit record, stale data can be written back to the filesystem as metadata in the recovery phase. To avoid this, when we failed to write out metadata buffers, abort the journal before writing the commit record. We can also avoid this kind of corruption by using the journal checksum feature because it can detect invalid metadata blocks in the journal and avoid them from being replayed. So we don't need to care about asynchronous commit record writeout with a checksum. Signed-off-by: Hidehiro Kawai Signed-off-by: Theodore Ts'o --- fs/jbd2/commit.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 0d3814a35ed1..78e4da934121 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -783,6 +783,9 @@ wait_for_iobuf: /* AKPM: bforget here */ } + if (err) + jbd2_journal_abort(journal, err); + jbd_debug(3, "JBD: commit phase 5\n"); if (!JBD2_HAS_INCOMPAT_FEATURE(journal, -- cgit v1.2.3 From 7ad7445f60fe4d46c4c9d2a9463db180d2a3b270 Mon Sep 17 00:00:00 2001 From: Hidehiro Kawai Date: Fri, 10 Oct 2008 20:29:31 -0400 Subject: jbd2: don't dirty original metadata buffer on abort Currently, original metadata buffers are dirtied when they are unfiled whether the journal has aborted or not. Eventually these buffers will be written-back to the filesystem by pdflush. This means some metadata buffers are written to the filesystem without journaling if the journal aborts. So if both journal abort and system crash happen at the same time, the filesystem would become inconsistent state. Additionally, replaying journaled metadata can overwrite the latest metadata on the filesystem partly. Because, if the journal gets aborted, journaled metadata are preserved and replayed during the next mount not to lose uncheckpointed metadata. This would also break the consistency of the filesystem. This patch prevents original metadata buffers from being dirtied on abort by clearing BH_JBDDirty flag from those buffers. Thus, no metadata buffers are written to the filesystem without journaling. Signed-off-by: Hidehiro Kawai Signed-off-by: Theodore Ts'o --- fs/jbd2/commit.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 78e4da934121..849f10496cea 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -504,9 +504,10 @@ 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_journal_refile_buffer(journal, jh); /* If that was the last one, we need to clean up @@ -884,6 +885,8 @@ restart_loop: if (buffer_jbddirty(bh)) { JBUFFER_TRACE(jh, "add to new checkpointing trans"); __jbd2_journal_insert_checkpoint(jh, commit_transaction); + if (is_journal_aborted(journal)) + clear_buffer_jbddirty(bh); JBUFFER_TRACE(jh, "refile for checkpoint writeback"); __jbd2_journal_refile_buffer(jh); jbd_unlock_bh_state(bh); -- cgit v1.2.3 From 5bf5683a33f3584da6eced480967c4f7e11515a8 Mon Sep 17 00:00:00 2001 From: Hidehiro Kawai Date: Fri, 10 Oct 2008 22:12:43 -0400 Subject: ext4: add an option to control error handling on file data If the journal doesn't abort when it gets an IO error in file data blocks, the file data corruption will spread silently. Because most of applications and commands do buffered writes without fsync(), they don't notice the IO error. It's scary for mission critical systems. On the other hand, if the journal aborts whenever it gets an IO error in file data blocks, the system will easily become inoperable. So this patch introduces a filesystem option to determine whether it aborts the journal or just call printk() when it gets an IO error in file data. If you mount an ext4 fs with data_err=abort option, it aborts on file data write error. If you mount it with data_err=ignore, it doesn't abort, just call printk(). data_err=ignore is the default. Here is the corresponding patch of the ext3 version: http://kerneltrap.org/mailarchive/linux-kernel/2008/9/9/3239374 Signed-off-by: Hidehiro Kawai Signed-off-by: Theodore Ts'o --- Documentation/filesystems/ext4.txt | 5 +++++ fs/ext4/ext4.h | 2 ++ fs/ext4/super.c | 16 ++++++++++++++++ fs/jbd2/commit.c | 2 ++ include/linux/jbd2.h | 3 +++ 5 files changed, 28 insertions(+) (limited to 'fs/jbd2/commit.c') diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt index 74484e696405..eb154ef36c2a 100644 --- a/Documentation/filesystems/ext4.txt +++ b/Documentation/filesystems/ext4.txt @@ -223,6 +223,11 @@ errors=remount-ro(*) Remount the filesystem read-only on an error. errors=continue Keep going on a filesystem error. errors=panic Panic and halt the machine if an error occurs. +data_err=ignore(*) Just print an error message if an error occurs + in a file data buffer in ordered mode. +data_err=abort Abort the journal if an error occurs in a file + data buffer in ordered mode. + grpid Give objects the same group ID as their creator. bsdgroups diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index f46a513a5157..6690a41cdd9f 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -540,6 +540,8 @@ do { \ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ #define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */ +#define EXT4_MOUNT_DATA_ERR_ABORT 0x10000000 /* Abort on file data write */ + /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H #define clear_opt(o, opt) o &= ~EXT4_MOUNT_##opt diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 79bd3989e84f..014677b8e224 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -778,6 +778,9 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) seq_printf(seq, ",inode_readahead_blks=%u", sbi->s_inode_readahead_blks); + if (test_opt(sb, DATA_ERR_ABORT)) + seq_puts(seq, ",data_err=abort"); + ext4_show_quota_options(seq, sb); return 0; } @@ -907,6 +910,7 @@ enum { Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, Opt_journal_checksum, Opt_journal_async_commit, Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, + Opt_data_err_abort, Opt_data_err_ignore, Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, @@ -953,6 +957,8 @@ static match_table_t tokens = { {Opt_data_journal, "data=journal"}, {Opt_data_ordered, "data=ordered"}, {Opt_data_writeback, "data=writeback"}, + {Opt_data_err_abort, "data_err=abort"}, + {Opt_data_err_ignore, "data_err=ignore"}, {Opt_offusrjquota, "usrjquota="}, {Opt_usrjquota, "usrjquota=%s"}, {Opt_offgrpjquota, "grpjquota="}, @@ -1187,6 +1193,12 @@ static int parse_options(char *options, struct super_block *sb, sbi->s_mount_opt |= data_opt; } break; + case Opt_data_err_abort: + set_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; + case Opt_data_err_ignore: + clear_opt(sbi->s_mount_opt, DATA_ERR_ABORT); + break; #ifdef CONFIG_QUOTA case Opt_usrjquota: qtype = USRQUOTA; @@ -2535,6 +2547,10 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) journal->j_flags |= JBD2_BARRIER; else journal->j_flags &= ~JBD2_BARRIER; + if (test_opt(sb, DATA_ERR_ABORT)) + journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR; + else + journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR; spin_unlock(&journal->j_state_lock); } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 849f10496cea..0abe02c4242a 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -684,6 +684,8 @@ start_journal_io: 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; } diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index c9e7d781db31..d2e91ea998fd 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -967,6 +967,9 @@ struct journal_s #define JBD2_FLUSHED 0x008 /* The journal superblock has been flushed */ #define JBD2_LOADED 0x010 /* The journal superblock has been loaded */ #define JBD2_BARRIER 0x020 /* Use IDE barriers */ +#define JBD2_ABORT_ON_SYNCDATA_ERR 0x040 /* Abort the journal on file + * data write error in ordered + * mode */ /* * Function declarations for the journaling transaction and buffer -- cgit v1.2.3 From 3e624fc72fba09b6f999a9fbb87b64efccd38036 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Thu, 16 Oct 2008 20:00:24 -0400 Subject: ext4: Replace hackish ext4_mb_poll_new_transaction with commit callback The multiblock allocator needs to be able to release blocks (and issue a blkdev discard request) when the transaction which freed those blocks is committed. Previously this was done via a polling mechanism when blocks are allocated or freed. A much better way of doing things is to create a jbd2 callback function and attaching the list of blocks to be freed directly to the transaction structure. Signed-off-by: "Theodore Ts'o" --- fs/ext4/ext4_sb.h | 3 -- fs/ext4/mballoc.c | 85 +++++++++------------------------------------------ fs/ext4/mballoc.h | 3 +- fs/jbd2/commit.c | 3 ++ fs/jbd2/transaction.c | 1 + include/linux/jbd2.h | 9 ++++++ 6 files changed, 29 insertions(+), 75 deletions(-) (limited to 'fs/jbd2/commit.c') diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h index 6a0b40d43264..445fde603df8 100644 --- a/fs/ext4/ext4_sb.h +++ b/fs/ext4/ext4_sb.h @@ -99,9 +99,6 @@ struct ext4_sb_info { struct inode *s_buddy_cache; long s_blocks_reserved; spinlock_t s_reserve_lock; - struct list_head s_active_transaction; - struct list_head s_closed_transaction; - struct list_head s_committed_transaction; spinlock_t s_md_lock; tid_t s_last_transaction; unsigned short *s_mb_offsets, *s_mb_maxs; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index da1da1fe2b14..dfe17a134052 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2523,9 +2523,6 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) } spin_lock_init(&sbi->s_md_lock); - INIT_LIST_HEAD(&sbi->s_active_transaction); - INIT_LIST_HEAD(&sbi->s_closed_transaction); - INIT_LIST_HEAD(&sbi->s_committed_transaction); spin_lock_init(&sbi->s_bal_lock); sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN; @@ -2554,6 +2551,8 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery) ext4_mb_init_per_dev_proc(sb); ext4_mb_history_init(sb); + sbi->s_journal->j_commit_callback = release_blocks_on_commit; + printk(KERN_INFO "EXT4-fs: mballoc enabled\n"); return 0; } @@ -2583,15 +2582,6 @@ int ext4_mb_release(struct super_block *sb) struct ext4_group_info *grinfo; struct ext4_sb_info *sbi = EXT4_SB(sb); - /* release freed, non-committed blocks */ - spin_lock(&sbi->s_md_lock); - list_splice_init(&sbi->s_closed_transaction, - &sbi->s_committed_transaction); - list_splice_init(&sbi->s_active_transaction, - &sbi->s_committed_transaction); - spin_unlock(&sbi->s_md_lock); - ext4_mb_free_committed_blocks(sb); - if (sbi->s_group_info) { for (i = 0; i < sbi->s_groups_count; i++) { grinfo = ext4_get_group_info(sb, i); @@ -2645,36 +2635,25 @@ int ext4_mb_release(struct super_block *sb) return 0; } -static noinline_for_stack void -ext4_mb_free_committed_blocks(struct super_block *sb) +/* + * This function is called by the jbd2 layer once the commit has finished, + * so we know we can free the blocks that were released with that commit. + */ +static void release_blocks_on_commit(journal_t *journal, transaction_t *txn) { + struct super_block *sb = journal->j_private; struct ext4_buddy e4b; struct ext4_group_info *db; - struct ext4_sb_info *sbi = EXT4_SB(sb); int err, count = 0, count2 = 0; struct ext4_free_data *entry; ext4_fsblk_t discard_block; + struct list_head *l, *ltmp; - if (list_empty(&sbi->s_committed_transaction)) - return; - - /* there is committed blocks to be freed yet */ - do { - /* get next array of blocks */ - entry = NULL; - spin_lock(&sbi->s_md_lock); - if (!list_empty(&sbi->s_committed_transaction)) { - entry = list_entry(sbi->s_committed_transaction.next, - struct ext4_free_data, list); - list_del(&entry->list); - } - spin_unlock(&sbi->s_md_lock); - - if (entry == NULL) - break; + list_for_each_safe(l, ltmp, &txn->t_private_list) { + entry = list_entry(l, struct ext4_free_data, list); mb_debug("gonna free %u blocks in group %lu (0x%p):", - entry->count, entry->group, entry); + entry->count, entry->group, entry); err = ext4_mb_load_buddy(sb, entry->group, &e4b); /* we expect to find existing buddy because it's pinned */ @@ -2706,7 +2685,7 @@ ext4_mb_free_committed_blocks(struct super_block *sb) kmem_cache_free(ext4_free_ext_cachep, entry); ext4_mb_release_desc(&e4b); - } while (1); + } mb_debug("freed %u blocks in %u structures\n", count, count2); } @@ -4348,8 +4327,6 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, goto out1; } - ext4_mb_poll_new_transaction(sb, handle); - *errp = ext4_mb_initialize_context(ac, ar); if (*errp) { ar->len = 0; @@ -4408,36 +4385,6 @@ out1: return block; } -static void ext4_mb_poll_new_transaction(struct super_block *sb, - handle_t *handle) -{ - struct ext4_sb_info *sbi = EXT4_SB(sb); - - if (sbi->s_last_transaction == handle->h_transaction->t_tid) - return; - - /* new transaction! time to close last one and free blocks for - * committed transaction. we know that only transaction can be - * active, so previos transaction can be being logged and we - * know that transaction before previous is known to be already - * logged. this means that now we may free blocks freed in all - * transactions before previous one. hope I'm clear enough ... */ - - spin_lock(&sbi->s_md_lock); - if (sbi->s_last_transaction != handle->h_transaction->t_tid) { - mb_debug("new transaction %lu, old %lu\n", - (unsigned long) handle->h_transaction->t_tid, - (unsigned long) sbi->s_last_transaction); - list_splice_init(&sbi->s_closed_transaction, - &sbi->s_committed_transaction); - list_splice_init(&sbi->s_active_transaction, - &sbi->s_closed_transaction); - sbi->s_last_transaction = handle->h_transaction->t_tid; - } - spin_unlock(&sbi->s_md_lock); - - ext4_mb_free_committed_blocks(sb); -} /* * We can merge two free data extents only if the physical blocks @@ -4531,9 +4478,9 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b, kmem_cache_free(ext4_free_ext_cachep, entry); } } - /* Add the extent to active_transaction list */ + /* Add the extent to transaction's private list */ spin_lock(&sbi->s_md_lock); - list_add(&new_entry->list, &sbi->s_active_transaction); + list_add(&new_entry->list, &handle->h_transaction->t_private_list); spin_unlock(&sbi->s_md_lock); ext4_unlock_group(sb, group); return 0; @@ -4562,8 +4509,6 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode, *freed = 0; - ext4_mb_poll_new_transaction(sb, handle); - sbi = EXT4_SB(sb); es = EXT4_SB(sb)->s_es; if (block < le32_to_cpu(es->s_first_data_block) || diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 94cb7b9fe3ee..b5dff1fff1e5 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -269,8 +269,6 @@ struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t); static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap, ext4_group_t group); -static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *); -static void ext4_mb_free_committed_blocks(struct super_block *); static void ext4_mb_return_to_preallocation(struct inode *inode, struct ext4_buddy *e4b, sector_t block, int count); @@ -278,6 +276,7 @@ static void ext4_mb_put_pa(struct ext4_allocation_context *, struct super_block *, struct ext4_prealloc_space *pa); static int ext4_mb_init_per_dev_proc(struct super_block *sb); static int ext4_mb_destroy_per_dev_proc(struct super_block *sb); +static void release_blocks_on_commit(journal_t *journal, transaction_t *txn); static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group) diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 0abe02c4242a..8b119e16aa36 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -995,6 +995,9 @@ restart_loop: } spin_unlock(&journal->j_list_lock); + if (journal->j_commit_callback) + journal->j_commit_callback(journal, commit_transaction); + trace_mark(jbd2_end_commit, "dev %s transaction %d head %d", journal->j_devname, commit_transaction->t_tid, journal->j_tail_sequence); diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index e5d540588fa9..39b7805a599a 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -52,6 +52,7 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) transaction->t_expires = jiffies + journal->j_commit_interval; spin_lock_init(&transaction->t_handle_lock); INIT_LIST_HEAD(&transaction->t_inode_list); + INIT_LIST_HEAD(&transaction->t_private_list); /* Set up the commit timer for the new transaction. */ journal->j_commit_timer.expires = round_jiffies(transaction->t_expires); diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h index 463d6f10b64f..c7d106ef22e2 100644 --- a/include/linux/jbd2.h +++ b/include/linux/jbd2.h @@ -641,6 +641,11 @@ struct transaction_s */ int t_handle_count; + /* + * For use by the filesystem to store fs-specific data + * structures associated with the transaction + */ + struct list_head t_private_list; }; struct transaction_run_stats_s { @@ -935,6 +940,10 @@ struct journal_s pid_t j_last_sync_writer; + /* This function is called when a transaction is closed */ + void (*j_commit_callback)(journal_t *, + transaction_t *); + /* * Journal statistics */ -- cgit v1.2.3