writeback: add tracepoints for cgroup foreign writebacks

cgroup foreign inode handling has quite a bit of heuristics and
internal states which sometimes makes it difficult to understand
what's going on.  Add tracepoints to improve visibility.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

3 files changed, 133 insertions, 0 deletions

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 658dc16c9e6d..8aaa7eec7b74 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -389,6 +389,8 @@ static void inode_switch_wbs_work_fn(struct work_struct *work)
 	if (unlikely(inode->i_state & I_FREEING))
 		goto skip_switch;
 
+	trace_inode_switch_wbs(inode, old_wb, new_wb);
+
 	/*
 	 * Count and transfer stats.  Note that PAGECACHE_TAG_DIRTY points
 	 * to possibly dirty pages while PAGECACHE_TAG_WRITEBACK points to
@@ -673,6 +675,9 @@ void wbc_detach_inode(struct writeback_control *wbc)
 		if (wbc->wb_id != max_id)
 			history |= (1U << slots) - 1;
 
+		if (history)
+			trace_inode_foreign_history(inode, wbc, history);
+
 		/*
 		 * Switch if the current wb isn't the consistent winner.
 		 * If there are multiple closely competing dirtiers, the
diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h
index aa7f3aeac740..3dc9fb9e7c78 100644
--- a/include/trace/events/writeback.h
+++ b/include/trace/events/writeback.h
@@ -176,6 +176,129 @@ static inline unsigned int __trace_wbc_assign_cgroup(struct writeback_control *w
 #endif	/* CONFIG_CGROUP_WRITEBACK */
 #endif	/* CREATE_TRACE_POINTS */
 
+#ifdef CONFIG_CGROUP_WRITEBACK
+TRACE_EVENT(inode_foreign_history,
+
+	TP_PROTO(struct inode *inode, struct writeback_control *wbc,
+		 unsigned int history),
+
+	TP_ARGS(inode, wbc, history),
+
+	TP_STRUCT__entry(
+		__array(char,		name, 32)
+		__field(unsigned long,	ino)
+		__field(unsigned int,	cgroup_ino)
+		__field(unsigned int,	history)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name, dev_name(inode_to_bdi(inode)->dev), 32);
+		__entry->ino		= inode->i_ino;
+		__entry->cgroup_ino	= __trace_wbc_assign_cgroup(wbc);
+		__entry->history	= history;
+	),
+
+	TP_printk("bdi %s: ino=%lu cgroup_ino=%u history=0x%x",
+		__entry->name,
+		__entry->ino,
+		__entry->cgroup_ino,
+		__entry->history
+	)
+);
+
+TRACE_EVENT(inode_switch_wbs,
+
+	TP_PROTO(struct inode *inode, struct bdi_writeback *old_wb,
+		 struct bdi_writeback *new_wb),
+
+	TP_ARGS(inode, old_wb, new_wb),
+
+	TP_STRUCT__entry(
+		__array(char,		name, 32)
+		__field(unsigned long,	ino)
+		__field(unsigned int,	old_cgroup_ino)
+		__field(unsigned int,	new_cgroup_ino)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name,	dev_name(old_wb->bdi->dev), 32);
+		__entry->ino		= inode->i_ino;
+		__entry->old_cgroup_ino	= __trace_wb_assign_cgroup(old_wb);
+		__entry->new_cgroup_ino	= __trace_wb_assign_cgroup(new_wb);
+	),
+
+	TP_printk("bdi %s: ino=%lu old_cgroup_ino=%u new_cgroup_ino=%u",
+		__entry->name,
+		__entry->ino,
+		__entry->old_cgroup_ino,
+		__entry->new_cgroup_ino
+	)
+);
+
+TRACE_EVENT(track_foreign_dirty,
+
+	TP_PROTO(struct page *page, struct bdi_writeback *wb),
+
+	TP_ARGS(page, wb),
+
+	TP_STRUCT__entry(
+		__array(char,		name, 32)
+		__field(u64,		bdi_id)
+		__field(unsigned long,	ino)
+		__field(unsigned int,	memcg_id)
+		__field(unsigned int,	cgroup_ino)
+		__field(unsigned int,	page_cgroup_ino)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name,	dev_name(wb->bdi->dev), 32);
+		__entry->bdi_id		= wb->bdi->id;
+		__entry->ino		= page->mapping->host->i_ino;
+		__entry->memcg_id	= wb->memcg_css->id;
+		__entry->cgroup_ino	= __trace_wb_assign_cgroup(wb);
+		__entry->page_cgroup_ino = page->mem_cgroup->css.cgroup->kn->id.ino;
+	),
+
+	TP_printk("bdi %s[%llu]: ino=%lu memcg_id=%u cgroup_ino=%u page_cgroup_ino=%u",
+		__entry->name,
+		__entry->bdi_id,
+		__entry->ino,
+		__entry->memcg_id,
+		__entry->cgroup_ino,
+		__entry->page_cgroup_ino
+	)
+);
+
+TRACE_EVENT(flush_foreign,
+
+	TP_PROTO(struct bdi_writeback *wb, unsigned int frn_bdi_id,
+		 unsigned int frn_memcg_id),
+
+	TP_ARGS(wb, frn_bdi_id, frn_memcg_id),
+
+	TP_STRUCT__entry(
+		__array(char,		name, 32)
+		__field(unsigned int,	cgroup_ino)
+		__field(unsigned int,	frn_bdi_id)
+		__field(unsigned int,	frn_memcg_id)
+	),
+
+	TP_fast_assign(
+		strncpy(__entry->name,	dev_name(wb->bdi->dev), 32);
+		__entry->cgroup_ino	= __trace_wb_assign_cgroup(wb);
+		__entry->frn_bdi_id	= frn_bdi_id;
+		__entry->frn_memcg_id	= frn_memcg_id;
+	),
+
+	TP_printk("bdi %s: cgroup_ino=%u frn_bdi_id=%u frn_memcg_id=%u",
+		__entry->name,
+		__entry->cgroup_ino,
+		__entry->frn_bdi_id,
+		__entry->frn_memcg_id
+	)
+);
+#endif
+
 DECLARE_EVENT_CLASS(writeback_write_inode_template,
 
 	TP_PROTO(struct inode *inode, struct writeback_control *wbc),
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 89b65f5ca634..4390994e8be9 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -4066,6 +4066,8 @@ static int mem_cgroup_oom_control_write(struct cgroup_subsys_state *css,
 
 #ifdef CONFIG_CGROUP_WRITEBACK
 
+#include <trace/events/writeback.h>
+
 static int memcg_wb_domain_init(struct mem_cgroup *memcg, gfp_t gfp)
 {
 	return wb_domain_init(&memcg->cgwb_domain, gfp);
@@ -4203,6 +4205,8 @@ void mem_cgroup_track_foreign_dirty_slowpath(struct page *page,
 	int oldest = -1;
 	int i;
 
+	trace_track_foreign_dirty(page, wb);
+
 	/*
 	 * Pick the slot to use.  If there is already a slot for @wb, keep
 	 * using it.  If not replace the oldest one which isn't being
@@ -4263,6 +4267,7 @@ void mem_cgroup_flush_foreign(struct bdi_writeback *wb)
 		if (time_after64(frn->at, now - intv) &&
 		    atomic_read(&frn->done.cnt) == 1) {
 			frn->at = 0;
+			trace_flush_foreign(wb, frn->bdi_id, frn->memcg_id);
 			cgroup_writeback_by_id(frn->bdi_id, frn->memcg_id, 0,
 					       WB_REASON_FOREIGN_FLUSH,
 					       &frn->done);