writeback: add more tracepoints

Add tracepoints for page dirtying, writeback_single_inode start, inode
dirtying and writeback.  For the latter two inode events, a pair of
events are defined to denote start and end of the operations (the
starting one has _start suffix and the one w/o suffix happens after
the operation is complete).  These inode ops are FS specific and can
be non-trivial and having enclosing tracepoints is useful for external
tracers.

This is part of tracepoint additions to improve visiblity into
dirtying / writeback operations for io tracer and userland.

v2: writeback_dirty_inode[_start] TPs may be called for files on
    pseudo FSes w/ unregistered bdi.  Check whether bdi->dev is %NULL
    before dereferencing.

v3: buffer dirtying moved to a block TP.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

fs/fs-writeback.c

@@ -318,8 +318,14 @@ static void queue_io(struct bdi_writeback *wb, struct wb_writeback_work *work)
 
 static int write_inode(struct inode *inode, struct writeback_control *wbc)
 {
-	if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode))
-		return inode->i_sb->s_op->write_inode(inode, wbc);
+	int ret;
+
+	if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode)) {
+		trace_writeback_write_inode_start(inode, wbc);
+		ret = inode->i_sb->s_op->write_inode(inode, wbc);
+		trace_writeback_write_inode(inode, wbc);
+		return ret;
+	}
 	return 0;
 }
 
@@ -450,6 +456,8 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 
 	WARN_ON(!(inode->i_state & I_SYNC));
 
+	trace_writeback_single_inode_start(inode, wbc, nr_to_write);
+
 	ret = do_writepages(mapping, wbc);
 
 	/*
@@ -1150,8 +1158,12 @@ void __mark_inode_dirty(struct inode *inode, int flags)
 	 * dirty the inode itself
 	 */
 	if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
+		trace_writeback_dirty_inode_start(inode, flags);
+
 		if (sb->s_op->dirty_inode)
 			sb->s_op->dirty_inode(inode, flags);
+
+		trace_writeback_dirty_inode(inode, flags);
 	}
 
 	/*

include/trace/events/writeback.h

@@ -32,6 +32,115 @@
 
 struct wb_writeback_work;
 
+TRACE_EVENT(writeback_dirty_page,
+
+	TP_PROTO(struct page *page, struct address_space *mapping),
+
+	TP_ARGS(page, mapping),
+
+	TP_STRUCT__entry (
+		__array(char, name, 32)
+		__field(unsigned long, ino)
+		__field(pgoff_t, index)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name,
+			mapping ? dev_name(mapping->backing_dev_info->dev) : "(unknown)", 32);
+		__entry->ino = mapping ? mapping->host->i_ino : 0;
+		__entry->index = page->index;
+	),
+
+	TP_printk("bdi %s: ino=%lu index=%lu",
+		__entry->name,
+		__entry->ino,
+		__entry->index
+	)
+);
+
+DECLARE_EVENT_CLASS(writeback_dirty_inode_template,
+
+	TP_PROTO(struct inode *inode, int flags),
+
+	TP_ARGS(inode, flags),
+
+	TP_STRUCT__entry (
+		__array(char, name, 32)
+		__field(unsigned long, ino)
+		__field(unsigned long, flags)
+	),
+
+	TP_fast_assign(
+		struct backing_dev_info *bdi = inode->i_mapping->backing_dev_info;
+
+		/* may be called for files on pseudo FSes w/ unregistered bdi */
+		strncpy(__entry->name,
+			bdi->dev ? dev_name(bdi->dev) : "(unknown)", 32);
+		__entry->ino		= inode->i_ino;
+		__entry->flags		= flags;
+	),
+
+	TP_printk("bdi %s: ino=%lu flags=%s",
+		__entry->name,
+		__entry->ino,
+		show_inode_state(__entry->flags)
+	)
+);
+
+DEFINE_EVENT(writeback_dirty_inode_template, writeback_dirty_inode_start,
+
+	TP_PROTO(struct inode *inode, int flags),
+
+	TP_ARGS(inode, flags)
+);
+
+DEFINE_EVENT(writeback_dirty_inode_template, writeback_dirty_inode,
+
+	TP_PROTO(struct inode *inode, int flags),
+
+	TP_ARGS(inode, flags)
+);
+
+DECLARE_EVENT_CLASS(writeback_write_inode_template,
+
+	TP_PROTO(struct inode *inode, struct writeback_control *wbc),
+
+	TP_ARGS(inode, wbc),
+
+	TP_STRUCT__entry (
+		__array(char, name, 32)
+		__field(unsigned long, ino)
+		__field(int, sync_mode)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name,
+			dev_name(inode->i_mapping->backing_dev_info->dev), 32);
+		__entry->ino		= inode->i_ino;
+		__entry->sync_mode	= wbc->sync_mode;
+	),
+
+	TP_printk("bdi %s: ino=%lu sync_mode=%d",
+		__entry->name,
+		__entry->ino,
+		__entry->sync_mode
+	)
+);
+
+DEFINE_EVENT(writeback_write_inode_template, writeback_write_inode_start,
+
+	TP_PROTO(struct inode *inode, struct writeback_control *wbc),
+
+	TP_ARGS(inode, wbc)
+);
+
+DEFINE_EVENT(writeback_write_inode_template, writeback_write_inode,
+
+	TP_PROTO(struct inode *inode, struct writeback_control *wbc),
+
+	TP_ARGS(inode, wbc)
+);
+
 DECLARE_EVENT_CLASS(writeback_work_class,
 	TP_PROTO(struct backing_dev_info *bdi, struct wb_writeback_work *work),
 	TP_ARGS(bdi, work),
@@ -479,6 +588,13 @@ DECLARE_EVENT_CLASS(writeback_single_inode_template,
 	)
 );
 
+DEFINE_EVENT(writeback_single_inode_template, writeback_single_inode_start,
+	TP_PROTO(struct inode *inode,
+		 struct writeback_control *wbc,
+		 unsigned long nr_to_write),
+	TP_ARGS(inode, wbc, nr_to_write)
+);
+
 DEFINE_EVENT(writeback_single_inode_template, writeback_single_inode,
 	TP_PROTO(struct inode *inode,
 		 struct writeback_control *wbc,

mm/page-writeback.c

@@ -1982,6 +1982,8 @@ int __set_page_dirty_no_writeback(struct page *page)
  */
 void account_page_dirtied(struct page *page, struct address_space *mapping)
 {
+	trace_writeback_dirty_page(page, mapping);
+
 	if (mapping_cap_account_dirty(mapping)) {
 		__inc_zone_page_state(page, NR_FILE_DIRTY);
 		__inc_zone_page_state(page, NR_DIRTIED);