block: add iostat counters for flush requests

Requests that triggers flushing volatile writeback cache to disk (barriers)
have significant effect to overall performance.

Block layer has sophisticated engine for combining several flush requests
into one. But there is no statistics for actual flushes executed by disk.
Requests which trigger flushes usually are barriers - zero-size writes.

This patch adds two iostat counters into /sys/class/block/$dev/stat and
/proc/diskstats - count of completed flush requests and their total time.

Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

3 files changed, 25 insertions, 5 deletions

diff --git a/block/blk-flush.c b/block/blk-flush.c
index 1eec9cbe5a0a..1777346baf06 100644
--- a/block/blk-flush.c
+++ b/block/blk-flush.c
@@ -136,6 +136,17 @@ static void blk_flush_queue_rq(struct request *rq, bool add_front)
 	blk_mq_add_to_requeue_list(rq, add_front, true);
 }
 
+static void blk_account_io_flush(struct request *rq)
+{
+	struct hd_struct *part = &rq->rq_disk->part0;
+
+	part_stat_lock();
+	part_stat_inc(part, ios[STAT_FLUSH]);
+	part_stat_add(part, nsecs[STAT_FLUSH],
+		      ktime_get_ns() - rq->start_time_ns);
+	part_stat_unlock();
+}
+
 /**
  * blk_flush_complete_seq - complete flush sequence
  * @rq: PREFLUSH/FUA request being sequenced
@@ -185,7 +196,7 @@ static void blk_flush_complete_seq(struct request *rq,
 
 	case REQ_FSEQ_DONE:
 		/*
-		 * @rq was previously adjusted by blk_flush_issue() for
+		 * @rq was previously adjusted by blk_insert_flush() for
 		 * flush sequencing and may already have gone through the
 		 * flush data request completion path.  Restore @rq for
 		 * normal completion and end it.
@@ -212,6 +223,8 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error)
 	struct blk_flush_queue *fq = blk_get_flush_queue(q, flush_rq->mq_ctx);
 	struct blk_mq_hw_ctx *hctx;
 
+	blk_account_io_flush(flush_rq);
+
 	/* release the tag's ownership to the req cloned from */
 	spin_lock_irqsave(&fq->mq_flush_lock, flags);
 
diff --git a/block/genhd.c b/block/genhd.c
index 26b31fcae217..ff6268970ddc 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1385,7 +1385,9 @@ static int diskstats_show(struct seq_file *seqf, void *v)
 			   "%lu %lu %lu %u "
 			   "%lu %lu %lu %u "
 			   "%u %u %u "
-			   "%lu %lu %lu %u\n",
+			   "%lu %lu %lu %u "
+			   "%lu %u"
+			   "\n",
 			   MAJOR(part_devt(hd)), MINOR(part_devt(hd)),
 			   disk_name(gp, hd->partno, buf),
 			   part_stat_read(hd, ios[STAT_READ]),
@@ -1402,7 +1404,9 @@ static int diskstats_show(struct seq_file *seqf, void *v)
 			   part_stat_read(hd, ios[STAT_DISCARD]),
 			   part_stat_read(hd, merges[STAT_DISCARD]),
 			   part_stat_read(hd, sectors[STAT_DISCARD]),
-			   (unsigned int)part_stat_read_msecs(hd, STAT_DISCARD)
+			   (unsigned int)part_stat_read_msecs(hd, STAT_DISCARD),
+			   part_stat_read(hd, ios[STAT_FLUSH]),
+			   (unsigned int)part_stat_read_msecs(hd, STAT_FLUSH)
 			);
 	}
 	disk_part_iter_exit(&piter);
diff --git a/block/partition-generic.c b/block/partition-generic.c
index aee643ce13d1..3db8b73a96b1 100644
--- a/block/partition-generic.c
+++ b/block/partition-generic.c
@@ -127,7 +127,8 @@ ssize_t part_stat_show(struct device *dev,
 		"%8lu %8lu %8llu %8u "
 		"%8lu %8lu %8llu %8u "
 		"%8u %8u %8u "
-		"%8lu %8lu %8llu %8u"
+		"%8lu %8lu %8llu %8u "
+		"%8lu %8u"
 		"\n",
 		part_stat_read(p, ios[STAT_READ]),
 		part_stat_read(p, merges[STAT_READ]),
@@ -143,7 +144,9 @@ ssize_t part_stat_show(struct device *dev,
 		part_stat_read(p, ios[STAT_DISCARD]),
 		part_stat_read(p, merges[STAT_DISCARD]),
 		(unsigned long long)part_stat_read(p, sectors[STAT_DISCARD]),
-		(unsigned int)part_stat_read_msecs(p, STAT_DISCARD));
+		(unsigned int)part_stat_read_msecs(p, STAT_DISCARD),
+		part_stat_read(p, ios[STAT_FLUSH]),
+		(unsigned int)part_stat_read_msecs(p, STAT_FLUSH));
 }
 
 ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr,