nilfs2: add a tracepoint for tracking stage transition of segment construction

This patch adds a tracepoint for tracking stage transition of block
collection in segment construction.  With the tracepoint, we can analysis
the behavior of segment construction in depth.  It would be useful for
bottleneck detection and debugging, etc.

The tracepoint is created with the standard trace API of linux (like ext3,
ext4, f2fs and btrfs).  So we can analysis with existing tools easily.  Of
course, more detailed analysis will be possible if we can create nilfs
specific analysis tools.

Below is an example of event dump with Brendan Gregg's perf-tools
(https://github.com/brendangregg/perf-tools).  Time consumption between
each stage can be obtained.

$ sudo bin/tpoint nilfs2:nilfs2_collection_stage_transition
Tracing nilfs2:nilfs2_collection_stage_transition. Ctrl-C to end.
        segctord-14875 [003] ...1 28311.067794: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_INIT
        segctord-14875 [003] ...1 28311.068139: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_GC
        segctord-14875 [003] ...1 28311.068139: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_FILE
        segctord-14875 [003] ...1 28311.068486: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_IFILE
        segctord-14875 [003] ...1 28311.068540: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_CPFILE
        segctord-14875 [003] ...1 28311.068561: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_SUFILE
        segctord-14875 [003] ...1 28311.068565: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_DAT
        segctord-14875 [003] ...1 28311.068573: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_SR
        segctord-14875 [003] ...1 28311.068574: nilfs2_collection_stage_transition: sci = ffff8800ce6de000 stage = ST_DONE

For capturing transition correctly, this patch adds wrappers for the
member scnt of nilfs_cstage.  With this change, every transition of the
stage can produce trace event in a correct manner.

Signed-off-by: Hitoshi Mitake <mitake.hitoshi@lab.ntt.co.jp>
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

3 files changed, 103 insertions, 21 deletions

diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index c6abbad9b8e3..ef354043b87c 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -77,6 +77,36 @@ enum {
 	NILFS_ST_DONE,
 };
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/nilfs2.h>
+
+/*
+ * nilfs_sc_cstage_inc(), nilfs_sc_cstage_set(), nilfs_sc_cstage_get() are
+ * wrapper functions of stage count (nilfs_sc_info->sc_stage.scnt). Users of
+ * the variable must use them because transition of stage count must involve
+ * trace events (trace_nilfs2_collection_stage_transition).
+ *
+ * nilfs_sc_cstage_get() isn't required for the above purpose because it doesn't
+ * produce tracepoint events. It is provided just for making the intention
+ * clear.
+ */
+static inline void nilfs_sc_cstage_inc(struct nilfs_sc_info *sci)
+{
+	sci->sc_stage.scnt++;
+	trace_nilfs2_collection_stage_transition(sci);
+}
+
+static inline void nilfs_sc_cstage_set(struct nilfs_sc_info *sci, int next_scnt)
+{
+	sci->sc_stage.scnt = next_scnt;
+	trace_nilfs2_collection_stage_transition(sci);
+}
+
+static inline int nilfs_sc_cstage_get(struct nilfs_sc_info *sci)
+{
+	return sci->sc_stage.scnt;
+}
+
 /* State flags of collection */
 #define NILFS_CF_NODE		0x0001	/* Collecting node blocks */
 #define NILFS_CF_IFILE_STARTED	0x0002	/* IFILE stage has started */
@@ -1062,7 +1092,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 	size_t ndone;
 	int err = 0;
 
-	switch (sci->sc_stage.scnt) {
+	switch (nilfs_sc_cstage_get(sci)) {
 	case NILFS_ST_INIT:
 		/* Pre-processes */
 		sci->sc_stage.flags = 0;
@@ -1071,7 +1101,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 			sci->sc_nblk_inc = 0;
 			sci->sc_curseg->sb_sum.flags = NILFS_SS_LOGBGN;
 			if (mode == SC_LSEG_DSYNC) {
-				sci->sc_stage.scnt = NILFS_ST_DSYNC;
+				nilfs_sc_cstage_set(sci, NILFS_ST_DSYNC);
 				goto dsync_mode;
 			}
 		}
@@ -1079,10 +1109,10 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 		sci->sc_stage.dirty_file_ptr = NULL;
 		sci->sc_stage.gc_inode_ptr = NULL;
 		if (mode == SC_FLUSH_DAT) {
-			sci->sc_stage.scnt = NILFS_ST_DAT;
+			nilfs_sc_cstage_set(sci, NILFS_ST_DAT);
 			goto dat_stage;
 		}
-		sci->sc_stage.scnt++;  /* Fall through */
+		nilfs_sc_cstage_inc(sci);  /* Fall through */
 	case NILFS_ST_GC:
 		if (nilfs_doing_gc()) {
 			head = &sci->sc_gc_inodes;
@@ -1103,7 +1133,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 			}
 			sci->sc_stage.gc_inode_ptr = NULL;
 		}
-		sci->sc_stage.scnt++;  /* Fall through */
+		nilfs_sc_cstage_inc(sci);  /* Fall through */
 	case NILFS_ST_FILE:
 		head = &sci->sc_dirty_files;
 		ii = list_prepare_entry(sci->sc_stage.dirty_file_ptr, head,
@@ -1125,10 +1155,10 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 		}
 		sci->sc_stage.dirty_file_ptr = NULL;
 		if (mode == SC_FLUSH_FILE) {
-			sci->sc_stage.scnt = NILFS_ST_DONE;
+			nilfs_sc_cstage_set(sci, NILFS_ST_DONE);
 			return 0;
 		}
-		sci->sc_stage.scnt++;
+		nilfs_sc_cstage_inc(sci);
 		sci->sc_stage.flags |= NILFS_CF_IFILE_STARTED;
 		/* Fall through */
 	case NILFS_ST_IFILE:
@@ -1136,7 +1166,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 					      &nilfs_sc_file_ops);
 		if (unlikely(err))
 			break;
-		sci->sc_stage.scnt++;
+		nilfs_sc_cstage_inc(sci);
 		/* Creating a checkpoint */
 		err = nilfs_segctor_create_checkpoint(sci);
 		if (unlikely(err))
@@ -1147,7 +1177,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 					      &nilfs_sc_file_ops);
 		if (unlikely(err))
 			break;
-		sci->sc_stage.scnt++;  /* Fall through */
+		nilfs_sc_cstage_inc(sci);  /* Fall through */
 	case NILFS_ST_SUFILE:
 		err = nilfs_sufile_freev(nilfs->ns_sufile, sci->sc_freesegs,
 					 sci->sc_nfreesegs, &ndone);
@@ -1163,7 +1193,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 					      &nilfs_sc_file_ops);
 		if (unlikely(err))
 			break;
-		sci->sc_stage.scnt++;  /* Fall through */
+		nilfs_sc_cstage_inc(sci);  /* Fall through */
 	case NILFS_ST_DAT:
  dat_stage:
 		err = nilfs_segctor_scan_file(sci, nilfs->ns_dat,
@@ -1171,10 +1201,10 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 		if (unlikely(err))
 			break;
 		if (mode == SC_FLUSH_DAT) {
-			sci->sc_stage.scnt = NILFS_ST_DONE;
+			nilfs_sc_cstage_set(sci, NILFS_ST_DONE);
 			return 0;
 		}
-		sci->sc_stage.scnt++;  /* Fall through */
+		nilfs_sc_cstage_inc(sci);  /* Fall through */
 	case NILFS_ST_SR:
 		if (mode == SC_LSEG_SR) {
 			/* Appending a super root */
@@ -1184,7 +1214,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 		}
 		/* End of a logical segment */
 		sci->sc_curseg->sb_sum.flags |= NILFS_SS_LOGEND;
-		sci->sc_stage.scnt = NILFS_ST_DONE;
+		nilfs_sc_cstage_set(sci, NILFS_ST_DONE);
 		return 0;
 	case NILFS_ST_DSYNC:
  dsync_mode:
@@ -1197,7 +1227,7 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 		if (unlikely(err))
 			break;
 		sci->sc_curseg->sb_sum.flags |= NILFS_SS_LOGEND;
-		sci->sc_stage.scnt = NILFS_ST_DONE;
+		nilfs_sc_cstage_set(sci, NILFS_ST_DONE);
 		return 0;
 	case NILFS_ST_DONE:
 		return 0;
@@ -1442,7 +1472,8 @@ static int nilfs_segctor_collect(struct nilfs_sc_info *sci,
 			goto failed;
 
 		/* The current segment is filled up */
-		if (mode != SC_LSEG_SR || sci->sc_stage.scnt < NILFS_ST_CPFILE)
+		if (mode != SC_LSEG_SR ||
+		    nilfs_sc_cstage_get(sci) < NILFS_ST_CPFILE)
 			break;
 
 		nilfs_clear_logs(&sci->sc_segbufs);
@@ -1946,7 +1977,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 	struct the_nilfs *nilfs = sci->sc_super->s_fs_info;
 	int err;
 
-	sci->sc_stage.scnt = NILFS_ST_INIT;
+	nilfs_sc_cstage_set(sci, NILFS_ST_INIT);
 	sci->sc_cno = nilfs->ns_cno;
 
 	err = nilfs_segctor_collect_dirty_files(sci, nilfs);
@@ -1974,7 +2005,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 			goto failed;
 
 		/* Avoid empty segment */
-		if (sci->sc_stage.scnt == NILFS_ST_DONE &&
+		if (nilfs_sc_cstage_get(sci) == NILFS_ST_DONE &&
 		    nilfs_segbuf_empty(sci->sc_curseg)) {
 			nilfs_segctor_abort_construction(sci, nilfs, 1);
 			goto out;
@@ -1988,7 +2019,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 			nilfs_segctor_fill_in_file_bmap(sci);
 
 		if (mode == SC_LSEG_SR &&
-		    sci->sc_stage.scnt >= NILFS_ST_CPFILE) {
+		    nilfs_sc_cstage_get(sci) >= NILFS_ST_CPFILE) {
 			err = nilfs_segctor_fill_in_checkpoint(sci);
 			if (unlikely(err))
 				goto failed_to_write;
@@ -2007,7 +2038,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 		if (unlikely(err))
 			goto failed_to_write;
 
-		if (sci->sc_stage.scnt == NILFS_ST_DONE ||
+		if (nilfs_sc_cstage_get(sci) == NILFS_ST_DONE ||
 		    nilfs->ns_blocksize_bits != PAGE_CACHE_SHIFT) {
 			/*
 			 * At this point, we avoid double buffering
@@ -2020,7 +2051,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 			if (err)
 				goto failed_to_write;
 		}
-	} while (sci->sc_stage.scnt != NILFS_ST_DONE);
+	} while (nilfs_sc_cstage_get(sci) != NILFS_ST_DONE);
 
  out:
 	nilfs_segctor_drop_written_files(sci, nilfs);
diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h
index a48d6de1e02c..0408b9b2814b 100644
--- a/fs/nilfs2/segment.h
+++ b/fs/nilfs2/segment.h
@@ -67,7 +67,8 @@ struct nilfs_recovery_info {
 
 /**
  * struct nilfs_cstage - Context of collection stage
- * @scnt: Stage count
+ * @scnt: Stage count, must be accessed via wrappers:
+ *        nilfs_sc_cstage_inc(), nilfs_sc_cstage_set(), nilfs_sc_cstage_get()
  * @flags: State flags
  * @dirty_file_ptr: Pointer on dirty_files list, or inode of a target file
  * @gc_inode_ptr: Pointer on the list of gc-inodes
diff --git a/include/trace/events/nilfs2.h b/include/trace/events/nilfs2.h
new file mode 100644
index 000000000000..573da00a486d
--- /dev/null
+++ b/include/trace/events/nilfs2.h
@@ -0,0 +1,50 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM nilfs2
+
+#if !defined(_TRACE_NILFS2_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NILFS2_H
+
+#include <linux/tracepoint.h>
+
+struct nilfs_sc_info;
+
+#define show_collection_stage(type)					\
+	__print_symbolic(type,						\
+	{ NILFS_ST_INIT, "ST_INIT" },					\
+	{ NILFS_ST_GC, "ST_GC" },					\
+	{ NILFS_ST_FILE, "ST_FILE" },					\
+	{ NILFS_ST_IFILE, "ST_IFILE" },					\
+	{ NILFS_ST_CPFILE, "ST_CPFILE" },				\
+	{ NILFS_ST_SUFILE, "ST_SUFILE" },				\
+	{ NILFS_ST_DAT, "ST_DAT" },					\
+	{ NILFS_ST_SR, "ST_SR" },					\
+	{ NILFS_ST_DSYNC, "ST_DSYNC" },					\
+	{ NILFS_ST_DONE, "ST_DONE"})
+
+TRACE_EVENT(nilfs2_collection_stage_transition,
+
+	    TP_PROTO(struct nilfs_sc_info *sci),
+
+	    TP_ARGS(sci),
+
+	    TP_STRUCT__entry(
+		    __field(void *, sci)
+		    __field(int, stage)
+	    ),
+
+	    TP_fast_assign(
+			__entry->sci = sci;
+			__entry->stage = sci->sc_stage.scnt;
+		    ),
+
+	    TP_printk("sci = %p stage = %s",
+		      __entry->sci,
+		      show_collection_stage(__entry->stage))
+);
+
+#endif /* _TRACE_NILFS2_H */
+
+/* This part must be outside protection */
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE nilfs2
+#include <trace/define_trace.h>