Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: Add new ext4 inode tracepoints
  ext4: Don't call sb_issue_discard() in ext4_free_blocks()
  ext4: do not try to grab the s_umount semaphore in ext4_quota_off
  ext4: fix potential race when freeing ext4_io_page structures
  ext4: handle writeback of inodes which are being freed
  ext4: initialize the percpu counters before replaying the journal
  ext4: "ret" may be used uninitialized in ext4_lazyinit_thread()
  ext4: fix lazyinit hang after removing request

fs/ext4/ext4.h

@@ -177,7 +177,7 @@ struct mpage_da_data {
 
 struct ext4_io_page {
 	struct page	*p_page;
-	int		p_count;
+	atomic_t	p_count;
 };
 
 #define MAX_IO_PAGES 128
@@ -858,6 +858,7 @@ struct ext4_inode_info {
 	spinlock_t i_completed_io_lock;
 	/* current io_end structure for async DIO write*/
 	ext4_io_end_t *cur_aio_dio;
+	atomic_t i_ioend_count;	/* Number of outstanding io_end structs */
 
 	/*
 	 * Transactions that contain inode's metadata needed to complete
@@ -2060,6 +2061,7 @@ extern int ext4_move_extents(struct file *o_filp, struct file *d_filp,
 /* page-io.c */
 extern int __init ext4_init_pageio(void);
 extern void ext4_exit_pageio(void);
+extern void ext4_ioend_wait(struct inode *);
 extern void ext4_free_io_end(ext4_io_end_t *io);
 extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags);
 extern int ext4_end_io_nolock(ext4_io_end_t *io);

fs/ext4/inode.c

@@ -53,6 +53,7 @@
 static inline int ext4_begin_ordered_truncate(struct inode *inode,
 					      loff_t new_size)
 {
+	trace_ext4_begin_ordered_truncate(inode, new_size);
 	return jbd2_journal_begin_ordered_truncate(
 					EXT4_SB(inode->i_sb)->s_journal,
 					&EXT4_I(inode)->jinode,
@@ -178,6 +179,7 @@ void ext4_evict_inode(struct inode *inode)
 	handle_t *handle;
 	int err;
 
+	trace_ext4_evict_inode(inode);
 	if (inode->i_nlink) {
 		truncate_inode_pages(&inode->i_data, 0);
 		goto no_delete;
@@ -5647,6 +5649,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode)
 	int err, ret;
 
 	might_sleep();
+	trace_ext4_mark_inode_dirty(inode, _RET_IP_);
 	err = ext4_reserve_inode_write(handle, inode, &iloc);
 	if (ext4_handle_valid(handle) &&
 	    EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize &&

fs/ext4/mballoc.c

@@ -4640,8 +4640,6 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		 * with group lock held. generate_buddy look at
 		 * them with group lock_held
 		 */
-		if (test_opt(sb, DISCARD))
-			ext4_issue_discard(sb, block_group, bit, count);
 		ext4_lock_group(sb, block_group);
 		mb_clear_bits(bitmap_bh->b_data, bit, count);
 		mb_free_blocks(inode, &e4b, bit, count);

fs/ext4/page-io.c

@@ -32,8 +32,14 @@
 
 static struct kmem_cache *io_page_cachep, *io_end_cachep;
 
+#define WQ_HASH_SZ		37
+#define to_ioend_wq(v)	(&ioend_wq[((unsigned long)v) % WQ_HASH_SZ])
+static wait_queue_head_t ioend_wq[WQ_HASH_SZ];
+
 int __init ext4_init_pageio(void)
 {
+	int i;
+
 	io_page_cachep = KMEM_CACHE(ext4_io_page, SLAB_RECLAIM_ACCOUNT);
 	if (io_page_cachep == NULL)
 		return -ENOMEM;
@@ -42,6 +48,8 @@ int __init ext4_init_pageio(void)
 		kmem_cache_destroy(io_page_cachep);
 		return -ENOMEM;
 	}
+	for (i = 0; i < WQ_HASH_SZ; i++)
+		init_waitqueue_head(&ioend_wq[i]);
 
 	return 0;
 }
@@ -52,24 +60,37 @@ void ext4_exit_pageio(void)
 	kmem_cache_destroy(io_page_cachep);
 }
 
+void ext4_ioend_wait(struct inode *inode)
+{
+	wait_queue_head_t *wq = to_ioend_wq(inode);
+
+	wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_ioend_count) == 0));
+}
+
+static void put_io_page(struct ext4_io_page *io_page)
+{
+	if (atomic_dec_and_test(&io_page->p_count)) {
+		end_page_writeback(io_page->p_page);
+		put_page(io_page->p_page);
+		kmem_cache_free(io_page_cachep, io_page);
+	}
+}
+
 void ext4_free_io_end(ext4_io_end_t *io)
 {
 	int i;
+	wait_queue_head_t *wq;
 
 	BUG_ON(!io);
 	if (io->page)
 		put_page(io->page);
-	for (i = 0; i < io->num_io_pages; i++) {
-		if (--io->pages[i]->p_count == 0) {
-			struct page *page = io->pages[i]->p_page;
-
-			end_page_writeback(page);
-			put_page(page);
-			kmem_cache_free(io_page_cachep, io->pages[i]);
-		}
-	}
+	for (i = 0; i < io->num_io_pages; i++)
+		put_io_page(io->pages[i]);
 	io->num_io_pages = 0;
-	iput(io->inode);
+	wq = to_ioend_wq(io->inode);
+	if (atomic_dec_and_test(&EXT4_I(io->inode)->i_ioend_count) &&
+	    waitqueue_active(wq))
+		wake_up_all(wq);
 	kmem_cache_free(io_end_cachep, io);
 }
 
@@ -142,8 +163,8 @@ ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
 	io = kmem_cache_alloc(io_end_cachep, flags);
 	if (io) {
 		memset(io, 0, sizeof(*io));
-		io->inode = igrab(inode);
-		BUG_ON(!io->inode);
+		atomic_inc(&EXT4_I(inode)->i_ioend_count);
+		io->inode = inode;
 		INIT_WORK(&io->work, ext4_end_io_work);
 		INIT_LIST_HEAD(&io->list);
 	}
@@ -171,35 +192,15 @@ static void ext4_end_bio(struct bio *bio, int error)
 	struct workqueue_struct *wq;
 	struct inode *inode;
 	unsigned long flags;
-	ext4_fsblk_t err_block;
 	int i;
 
 	BUG_ON(!io_end);
-	inode = io_end->inode;
 	bio->bi_private = NULL;
 	bio->bi_end_io = NULL;
 	if (test_bit(BIO_UPTODATE, &bio->bi_flags))
 		error = 0;
-	err_block = bio->bi_sector >> (inode->i_blkbits - 9);
 	bio_put(bio);
 
-	if (!(inode->i_sb->s_flags & MS_ACTIVE)) {
-		pr_err("sb umounted, discard end_io request for inode %lu\n",
-			io_end->inode->i_ino);
-		ext4_free_io_end(io_end);
-		return;
-	}
-
-	if (error) {
-		io_end->flag |= EXT4_IO_END_ERROR;
-		ext4_warning(inode->i_sb, "I/O error writing to inode %lu "
-			     "(offset %llu size %ld starting block %llu)",
-			     inode->i_ino,
-			     (unsigned long long) io_end->offset,
-			     (long) io_end->size,
-			     (unsigned long long) err_block);
-	}
-
 	for (i = 0; i < io_end->num_io_pages; i++) {
 		struct page *page = io_end->pages[i]->p_page;
 		struct buffer_head *bh, *head;
@@ -236,13 +237,7 @@ static void ext4_end_bio(struct bio *bio, int error)
 			} while (bh != head);
 		}
 
-		if (--io_end->pages[i]->p_count == 0) {
-			struct page *page = io_end->pages[i]->p_page;
-
-			end_page_writeback(page);
-			put_page(page);
-			kmem_cache_free(io_page_cachep, io_end->pages[i]);
-		}
+		put_io_page(io_end->pages[i]);
 
 		/*
 		 * If this is a partial write which happened to make
@@ -254,8 +249,19 @@ static void ext4_end_bio(struct bio *bio, int error)
 		if (!partial_write)
 			SetPageUptodate(page);
 	}
-
 	io_end->num_io_pages = 0;
+	inode = io_end->inode;
+
+	if (error) {
+		io_end->flag |= EXT4_IO_END_ERROR;
+		ext4_warning(inode->i_sb, "I/O error writing to inode %lu "
+			     "(offset %llu size %ld starting block %llu)",
+			     inode->i_ino,
+			     (unsigned long long) io_end->offset,
+			     (long) io_end->size,
+			     (unsigned long long)
+			     bio->bi_sector >> (inode->i_blkbits - 9));
+	}
 
 	/* Add the io_end to per-inode completed io list*/
 	spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags);
@@ -305,7 +311,6 @@ static int io_submit_init(struct ext4_io_submit *io,
 	bio->bi_private = io->io_end = io_end;
 	bio->bi_end_io = ext4_end_bio;
 
-	io_end->inode = inode;
 	io_end->offset = (page->index << PAGE_CACHE_SHIFT) + bh_offset(bh);
 
 	io->io_bio = bio;
@@ -360,7 +365,7 @@ static int io_submit_add_bh(struct ext4_io_submit *io,
 	if ((io_end->num_io_pages == 0) ||
 	    (io_end->pages[io_end->num_io_pages-1] != io_page)) {
 		io_end->pages[io_end->num_io_pages++] = io_page;
-		io_page->p_count++;
+		atomic_inc(&io_page->p_count);
 	}
 	return 0;
 }
@@ -389,7 +394,7 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
 		return -ENOMEM;
 	}
 	io_page->p_page = page;
-	io_page->p_count = 0;
+	atomic_set(&io_page->p_count, 1);
 	get_page(page);
 
 	for (bh = head = page_buffers(page), block_start = 0;
@@ -421,10 +426,6 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
 	 * PageWriteback bit from the page to prevent the system from
 	 * wedging later on.
 	 */
-	if (io_page->p_count == 0) {
-		put_page(page);
-		end_page_writeback(page);
-		kmem_cache_free(io_page_cachep, io_page);
-	}
+	put_io_page(io_page);
 	return ret;
 }

fs/ext4/super.c

@@ -828,12 +828,22 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei->cur_aio_dio = NULL;
 	ei->i_sync_tid = 0;
 	ei->i_datasync_tid = 0;
+	atomic_set(&ei->i_ioend_count, 0);
 
 	return &ei->vfs_inode;
 }
 
+static int ext4_drop_inode(struct inode *inode)
+{
+	int drop = generic_drop_inode(inode);
+
+	trace_ext4_drop_inode(inode, drop);
+	return drop;
+}
+
 static void ext4_destroy_inode(struct inode *inode)
 {
+	ext4_ioend_wait(inode);
 	if (!list_empty(&(EXT4_I(inode)->i_orphan))) {
 		ext4_msg(inode->i_sb, KERN_ERR,
 			 "Inode %lu (%p): orphan list check failed!",
@@ -1173,6 +1183,7 @@ static const struct super_operations ext4_sops = {
 	.destroy_inode	= ext4_destroy_inode,
 	.write_inode	= ext4_write_inode,
 	.dirty_inode	= ext4_dirty_inode,
+	.drop_inode	= ext4_drop_inode,
 	.evict_inode	= ext4_evict_inode,
 	.put_super	= ext4_put_super,
 	.sync_fs	= ext4_sync_fs,
@@ -1194,6 +1205,7 @@ static const struct super_operations ext4_nojournal_sops = {
 	.destroy_inode	= ext4_destroy_inode,
 	.write_inode	= ext4_write_inode,
 	.dirty_inode	= ext4_dirty_inode,
+	.drop_inode	= ext4_drop_inode,
 	.evict_inode	= ext4_evict_inode,
 	.write_super	= ext4_write_super,
 	.put_super	= ext4_put_super,
@@ -2699,7 +2711,6 @@ static int ext4_lazyinit_thread(void *arg)
 	struct ext4_li_request *elr;
 	unsigned long next_wakeup;
 	DEFINE_WAIT(wait);
-	int ret;
 
 	BUG_ON(NULL == eli);
 
@@ -2723,13 +2734,12 @@ static int ext4_lazyinit_thread(void *arg)
 			elr = list_entry(pos, struct ext4_li_request,
 					 lr_request);
 
-			if (time_after_eq(jiffies, elr->lr_next_sched))
-				ret = ext4_run_li_request(elr);
-
-			if (ret) {
-				ret = 0;
-				ext4_remove_li_request(elr);
-				continue;
+			if (time_after_eq(jiffies, elr->lr_next_sched)) {
+				if (ext4_run_li_request(elr) != 0) {
+					/* error, remove the lazy_init job */
+					ext4_remove_li_request(elr);
+					continue;
+				}
 			}
 
 			if (time_before(elr->lr_next_sched, next_wakeup))
@@ -2740,7 +2750,8 @@ static int ext4_lazyinit_thread(void *arg)
 		if (freezing(current))
 			refrigerator();
 
-		if (time_after_eq(jiffies, next_wakeup)) {
+		if ((time_after_eq(jiffies, next_wakeup)) ||
+		    (MAX_JIFFY_OFFSET == next_wakeup)) {
 			cond_resched();
 			continue;
 		}
@@ -3348,6 +3359,24 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 	spin_lock_init(&sbi->s_next_gen_lock);
 
+	err = percpu_counter_init(&sbi->s_freeblocks_counter,
+			ext4_count_free_blocks(sb));
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_freeinodes_counter,
+				ext4_count_free_inodes(sb));
+	}
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_dirs_counter,
+				ext4_count_dirs(sb));
+	}
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0);
+	}
+	if (err) {
+		ext4_msg(sb, KERN_ERR, "insufficient memory");
+		goto failed_mount3;
+	}
+
 	sbi->s_stripe = ext4_get_stripe_size(sbi);
 	sbi->s_max_writeback_mb_bump = 128;
 
@@ -3446,22 +3475,19 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	}
 	set_task_ioprio(sbi->s_journal->j_task, journal_ioprio);
 
-no_journal:
-	err = percpu_counter_init(&sbi->s_freeblocks_counter,
-				  ext4_count_free_blocks(sb));
-	if (!err)
-		err = percpu_counter_init(&sbi->s_freeinodes_counter,
-					  ext4_count_free_inodes(sb));
-	if (!err)
-		err = percpu_counter_init(&sbi->s_dirs_counter,
-					  ext4_count_dirs(sb));
-	if (!err)
-		err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0);
-	if (err) {
-		ext4_msg(sb, KERN_ERR, "insufficient memory");
-		goto failed_mount_wq;
-	}
+	/*
+	 * The journal may have updated the bg summary counts, so we
+	 * need to update the global counters.
+	 */
+	percpu_counter_set(&sbi->s_freeblocks_counter,
+			   ext4_count_free_blocks(sb));
+	percpu_counter_set(&sbi->s_freeinodes_counter,
+			   ext4_count_free_inodes(sb));
+	percpu_counter_set(&sbi->s_dirs_counter,
+			   ext4_count_dirs(sb));
+	percpu_counter_set(&sbi->s_dirtyblocks_counter, 0);
 
+no_journal:
 	EXT4_SB(sb)->dio_unwritten_wq = create_workqueue("ext4-dio-unwritten");
 	if (!EXT4_SB(sb)->dio_unwritten_wq) {
 		printk(KERN_ERR "EXT4-fs: failed to create DIO workqueue\n");
@@ -3611,10 +3637,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
 	}
-	percpu_counter_destroy(&sbi->s_freeblocks_counter);
-	percpu_counter_destroy(&sbi->s_freeinodes_counter);
-	percpu_counter_destroy(&sbi->s_dirs_counter);
-	percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
 failed_mount3:
 	if (sbi->s_flex_groups) {
 		if (is_vmalloc_addr(sbi->s_flex_groups))
@@ -3622,6 +3644,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		else
 			kfree(sbi->s_flex_groups);
 	}
+	percpu_counter_destroy(&sbi->s_freeblocks_counter);
+	percpu_counter_destroy(&sbi->s_freeinodes_counter);
+	percpu_counter_destroy(&sbi->s_dirs_counter);
+	percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
 failed_mount2:
 	for (i = 0; i < db_count; i++)
 		brelse(sbi->s_group_desc[i]);
@@ -3949,13 +3975,11 @@ static int ext4_commit_super(struct super_block *sb, int sync)
 	else
 		es->s_kbytes_written =
 			cpu_to_le64(EXT4_SB(sb)->s_kbytes_written);
-	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeblocks_counter))
-		ext4_free_blocks_count_set(es, percpu_counter_sum_positive(
-					&EXT4_SB(sb)->s_freeblocks_counter));
-	if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter))
-		es->s_free_inodes_count =
-			cpu_to_le32(percpu_counter_sum_positive(
-					&EXT4_SB(sb)->s_freeinodes_counter));
+	ext4_free_blocks_count_set(es, percpu_counter_sum_positive(
+					   &EXT4_SB(sb)->s_freeblocks_counter));
+	es->s_free_inodes_count =
+		cpu_to_le32(percpu_counter_sum_positive(
+				&EXT4_SB(sb)->s_freeinodes_counter));
 	sb->s_dirt = 0;
 	BUFFER_TRACE(sbh, "marking dirty");
 	mark_buffer_dirty(sbh);
@@ -4556,12 +4580,10 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,
 
 static int ext4_quota_off(struct super_block *sb, int type)
 {
-	/* Force all delayed allocation blocks to be allocated */
-	if (test_opt(sb, DELALLOC)) {
-		down_read(&sb->s_umount);
+	/* Force all delayed allocation blocks to be allocated.
+	 * Caller already holds s_umount sem */
+	if (test_opt(sb, DELALLOC))
 		sync_filesystem(sb);
-		up_read(&sb->s_umount);
-	}
 
 	return dquot_quota_off(sb, type);
 }

include/trace/events/ext4.h

@@ -98,6 +98,103 @@ TRACE_EVENT(ext4_allocate_inode,
 		  (unsigned long) __entry->dir, __entry->mode)
 );
 
+TRACE_EVENT(ext4_evict_inode,
+	TP_PROTO(struct inode *inode),
+
+	TP_ARGS(inode),
+
+	TP_STRUCT__entry(
+		__field(	int,   dev_major                )
+		__field(	int,   dev_minor                )
+		__field(	ino_t,	ino			)
+		__field(	int,	nlink			)
+	),
+
+	TP_fast_assign(
+		__entry->dev_major = MAJOR(inode->i_sb->s_dev);
+		__entry->dev_minor = MINOR(inode->i_sb->s_dev);
+		__entry->ino	= inode->i_ino;
+		__entry->nlink	= inode->i_nlink;
+	),
+
+	TP_printk("dev %d,%d ino %lu nlink %d",
+		  __entry->dev_major, __entry->dev_minor,
+		  (unsigned long) __entry->ino, __entry->nlink)
+);
+
+TRACE_EVENT(ext4_drop_inode,
+	TP_PROTO(struct inode *inode, int drop),
+
+	TP_ARGS(inode, drop),
+
+	TP_STRUCT__entry(
+		__field(	int,	dev_major		)
+		__field(	int,	dev_minor		)
+		__field(	ino_t,	ino			)
+		__field(	int,	drop			)
+	),
+
+	TP_fast_assign(
+		__entry->dev_major = MAJOR(inode->i_sb->s_dev);
+		__entry->dev_minor = MINOR(inode->i_sb->s_dev);
+		__entry->ino	= inode->i_ino;
+		__entry->drop	= drop;
+	),
+
+	TP_printk("dev %d,%d ino %lu drop %d",
+		  __entry->dev_major, __entry->dev_minor,
+		  (unsigned long) __entry->ino, __entry->drop)
+);
+
+TRACE_EVENT(ext4_mark_inode_dirty,
+	TP_PROTO(struct inode *inode, unsigned long IP),
+
+	TP_ARGS(inode, IP),
+
+	TP_STRUCT__entry(
+		__field(	int,	dev_major		)
+		__field(	int,	dev_minor		)
+		__field(	ino_t,	ino			)
+		__field(unsigned long,	ip			)
+	),
+
+	TP_fast_assign(
+		__entry->dev_major = MAJOR(inode->i_sb->s_dev);
+		__entry->dev_minor = MINOR(inode->i_sb->s_dev);
+		__entry->ino	= inode->i_ino;
+		__entry->ip	= IP;
+	),
+
+	TP_printk("dev %d,%d ino %lu caller %pF",
+		  __entry->dev_major, __entry->dev_minor,
+		  (unsigned long) __entry->ino, (void *)__entry->ip)
+);
+
+TRACE_EVENT(ext4_begin_ordered_truncate,
+	TP_PROTO(struct inode *inode, loff_t new_size),
+
+	TP_ARGS(inode, new_size),
+
+	TP_STRUCT__entry(
+		__field(	int,	dev_major		)
+		__field(	int,	dev_minor		)
+		__field(	ino_t,	ino			)
+		__field(	loff_t,	new_size		)
+	),
+
+	TP_fast_assign(
+		__entry->dev_major	= MAJOR(inode->i_sb->s_dev);
+		__entry->dev_minor	= MINOR(inode->i_sb->s_dev);
+		__entry->ino		= inode->i_ino;
+		__entry->new_size	= new_size;
+	),
+
+	TP_printk("dev %d,%d ino %lu new_size %lld",
+		  __entry->dev_major, __entry->dev_minor,
+		  (unsigned long) __entry->ino,
+		  (long long) __entry->new_size)
+);
+
 DECLARE_EVENT_CLASS(ext4__write_begin,
 
 	TP_PROTO(struct inode *inode, loff_t pos, unsigned int len,