[PATCH 3/4] block: add io_extra_stats node

From: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
To: axboe@kernel.dk
Cc: linux-block@vger.kernel.org, danil.kipnis@cloud.ionos.com,
	jinpu.wang@cloud.ionos.com,
	Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
Subject: [PATCH 3/4] block: add io_extra_stats node
Date: Wed, 27 Jan 2021 15:59:29 +0100	[thread overview]
Message-ID: <20210127145930.8826-4-guoqing.jiang@cloud.ionos.com> (raw)
In-Reply-To: <20210127145930.8826-1-guoqing.jiang@cloud.ionos.com>

Even we have introduced a Kconfig option (default N) to control the
accounting of additional data, but the option still could be enabled
occasionally while user doesn't care about the size and latency of io,
and they could suffer from the additional overhead. So introduce a
specific sysfs node to avoid such mistake.

Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
---
 Documentation/ABI/testing/sysfs-block |  9 +++++++++
 Documentation/block/queue-sysfs.rst   |  6 ++++++
 block/blk-sysfs.c                     | 10 ++++++++++
 include/linux/blkdev.h                |  6 ++++++
 4 files changed, 31 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-block b/Documentation/ABI/testing/sysfs-block
index 0ffb63469772..e1611c62a3e1 100644
--- a/Documentation/ABI/testing/sysfs-block
+++ b/Documentation/ABI/testing/sysfs-block
@@ -333,3 +333,12 @@ Description:
 		does not complete in this time then the block driver timeout
 		handler is invoked. That timeout handler can decide to retry
 		the request, to fail it or to start a device recovery strategy.
+
+What:		/sys/block/<disk>/queue/io_extra_stats
+Date:		January 2021
+Contact:	Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
+Description:
+		Indicates if people want to know the extra statistics (I/O
+		size and I/O latency) from /sys/block/<disk>/io_latency
+		and /sys/block/<disk>/io_size. The value is 0 by default,
+		set if the extra statistics are needed.
diff --git a/Documentation/block/queue-sysfs.rst b/Documentation/block/queue-sysfs.rst
index 2638d3446b79..14c4a4c7b9a9 100644
--- a/Documentation/block/queue-sysfs.rst
+++ b/Documentation/block/queue-sysfs.rst
@@ -99,6 +99,12 @@ iostats (RW)
 This file is used to control (on/off) the iostats accounting of the
 disk.
 
+io_extra_stats (RW)
+-------------------
+This file is used to control (on/off) the additional accounting of the
+io size and io latency of disk, and BLK_ADDITIONAL_DISKSTAT should be
+enabled if you want the additional accounting.
+
 logical_block_size (RO)
 -----------------------
 This is the logical block size of the device, in bytes.
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index b513f1683af0..87f174f32e9a 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -288,6 +288,9 @@ QUEUE_SYSFS_BIT_FNS(nonrot, NONROT, 1);
 QUEUE_SYSFS_BIT_FNS(random, ADD_RANDOM, 0);
 QUEUE_SYSFS_BIT_FNS(iostats, IO_STAT, 0);
 QUEUE_SYSFS_BIT_FNS(stable_writes, STABLE_WRITES, 0);
+#ifdef CONFIG_BLK_ADDITIONAL_DISKSTAT
+QUEUE_SYSFS_BIT_FNS(io_extra_stats, IO_EXTRA_STAT, 0);
+#endif
 #undef QUEUE_SYSFS_BIT_FNS
 
 static ssize_t queue_zoned_show(struct request_queue *q, char *page)
@@ -616,6 +619,10 @@ QUEUE_RW_ENTRY(queue_iostats, "iostats");
 QUEUE_RW_ENTRY(queue_random, "add_random");
 QUEUE_RW_ENTRY(queue_stable_writes, "stable_writes");
 
+#ifdef CONFIG_BLK_ADDITIONAL_DISKSTAT
+QUEUE_RW_ENTRY(queue_io_extra_stats, "io_extra_stats");
+#endif
+
 static struct attribute *queue_attrs[] = {
 	&queue_requests_entry.attr,
 	&queue_ra_entry.attr,
@@ -658,6 +665,9 @@ static struct attribute *queue_attrs[] = {
 	&queue_io_timeout_entry.attr,
 #ifdef CONFIG_BLK_DEV_THROTTLING_LOW
 	&blk_throtl_sample_time_entry.attr,
+#endif
+#ifdef CONFIG_BLK_ADDITIONAL_DISKSTAT
+	&queue_io_extra_stats_entry.attr,
 #endif
 	NULL,
 };
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 20f3706b6b2e..6cebc689c36a 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -621,6 +621,7 @@ struct request_queue {
 #define QUEUE_FLAG_RQ_ALLOC_TIME 27	/* record rq->alloc_time_ns */
 #define QUEUE_FLAG_HCTX_ACTIVE	28	/* at least one blk-mq hctx is active */
 #define QUEUE_FLAG_NOWAIT       29	/* device supports NOWAIT */
+#define QUEUE_FLAG_IO_EXTRA_STAT 30	/* extra IO accounting for size and latency */
 
 #define QUEUE_FLAG_MQ_DEFAULT	((1 << QUEUE_FLAG_IO_STAT) |		\
 				 (1 << QUEUE_FLAG_SAME_COMP) |		\
@@ -658,6 +659,11 @@ bool blk_queue_flag_test_and_set(unsigned int flag, struct request_queue *q);
 #else
 #define blk_queue_rq_alloc_time(q)	false
 #endif
+#ifdef CONFIG_BLK_ADDITIONAL_DISKSTAT
+#define blk_queue_io_extra_stat(q) test_bit(QUEUE_FLAG_IO_EXTRA_STAT, &(q)->queue_flags)
+#else
+#define blk_queue_io_extra_stat(q) false
+#endif
 
 #define blk_noretry_request(rq) \
 	((rq)->cmd_flags & (REQ_FAILFAST_DEV|REQ_FAILFAST_TRANSPORT| \
-- 
2.17.1

