linux-kernel - [PATCH v3 5/9] block: export API to get the number of bdev inflight IO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250506124903.2540268-6-yukuai1@huaweicloud.com>
Date: Tue,  6 May 2025 20:48:59 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: axboe@...nel.dk,
	agk@...hat.com,
	song@...nel.org,
	hch@....de,
	john.g.garry@...cle.com,
	hare@...e.de,
	xni@...hat.com,
	pmenzel@...gen.mpg.de
Cc: linux-block@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	dm-devel@...ts.linux.dev,
	linux-raid@...r.kernel.org,
	yukuai3@...wei.com,
	yukuai1@...weicloud.com,
	yi.zhang@...wei.com,
	yangerkun@...wei.com,
	johnny.chenyi@...wei.com
Subject: [PATCH v3 5/9] block: export API to get the number of bdev inflight IO

From: Yu Kuai <yukuai3@...wei.com>

- rename part_in_{flight, flight_rw} to bdev_count_{inflight, inflight_rw}
- export bdev_count_inflight, to fix a problem in mdraid that foreground
  IO can be starved by background sync IO in later patches

Signed-off-by: Yu Kuai <yukuai3@...wei.com>
Reviewed-by: Christoph Hellwig <hch@....de>
Reviewed-by: Hannes Reinecke <hare@...e.de>
---
 block/blk-core.c          |  2 +-
 block/blk.h               |  1 -
 block/genhd.c             | 22 ++++++++++++++++------
 include/linux/part_stat.h |  2 ++
 4 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index e8cc270a453f..b862c66018f2 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1018,7 +1018,7 @@ void update_io_ticks(struct block_device *part, unsigned long now, bool end)
 	stamp = READ_ONCE(part->bd_stamp);
 	if (unlikely(time_after(now, stamp)) &&
 	    likely(try_cmpxchg(&part->bd_stamp, &stamp, now)) &&
-	    (end || part_in_flight(part)))
+	    (end || bdev_count_inflight(part)))
 		__part_stat_add(part, io_ticks, now - stamp);
 
 	if (bdev_is_partition(part)) {
diff --git a/block/blk.h b/block/blk.h
index ff21f234b62f..c5a18f0fd3a9 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -421,7 +421,6 @@ void blk_apply_bdi_limits(struct backing_dev_info *bdi,
 int blk_dev_init(void);
 
 void update_io_ticks(struct block_device *part, unsigned long now, bool end);
-unsigned int part_in_flight(struct block_device *part);
 
 static inline void req_set_nomerge(struct request_queue *q, struct request *req)
 {
diff --git a/block/genhd.c b/block/genhd.c
index 2470099b492b..fdaeafddfc4c 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -125,7 +125,7 @@ static void part_stat_read_all(struct block_device *part,
 	}
 }
 
-static void part_in_flight_rw(struct block_device *part,
+static void bdev_count_inflight_rw(struct block_device *part,
 		unsigned int inflight[2], bool mq_driver)
 {
 	int cpu;
@@ -147,14 +147,24 @@ static void part_in_flight_rw(struct block_device *part,
 		inflight[WRITE] = 0;
 }
 
-unsigned int part_in_flight(struct block_device *part)
+/**
+ * bdev_count_inflight - get the number of inflight IOs for a block device.
+ *
+ * @part: the block device.
+ *
+ * Inflight here means started IO accounting, from bdev_start_io_acct() for
+ * bio-based block device, and from blk_account_io_start() for rq-based block
+ * device.
+ */
+unsigned int bdev_count_inflight(struct block_device *part)
 {
 	unsigned int inflight[2] = {0};
 
-	part_in_flight_rw(part, inflight, false);
+	bdev_count_inflight_rw(part, inflight, false);
 
 	return inflight[READ] + inflight[WRITE];
 }
+EXPORT_SYMBOL_GPL(bdev_count_inflight);
 
 /*
  * Can be deleted altogether. Later.
@@ -1004,7 +1014,7 @@ ssize_t part_stat_show(struct device *dev,
 	struct disk_stats stat;
 	unsigned int inflight;
 
-	inflight = part_in_flight(bdev);
+	inflight = bdev_count_inflight(bdev);
 	if (inflight) {
 		part_stat_lock();
 		update_io_ticks(bdev, jiffies, true);
@@ -1053,7 +1063,7 @@ ssize_t part_inflight_show(struct device *dev, struct device_attribute *attr,
 	struct request_queue *q = bdev_get_queue(bdev);
 	unsigned int inflight[2] = {0};
 
-	part_in_flight_rw(bdev, inflight, queue_is_mq(q));
+	bdev_count_inflight_rw(bdev, inflight, queue_is_mq(q));
 
 	return sysfs_emit(buf, "%8u %8u\n", inflight[READ], inflight[WRITE]);
 }
@@ -1308,7 +1318,7 @@ static int diskstats_show(struct seq_file *seqf, void *v)
 		if (bdev_is_partition(hd) && !bdev_nr_sectors(hd))
 			continue;
 
-		inflight = part_in_flight(hd);
+		inflight = bdev_count_inflight(hd);
 		if (inflight) {
 			part_stat_lock();
 			update_io_ticks(hd, jiffies, true);
diff --git a/include/linux/part_stat.h b/include/linux/part_stat.h
index c5e9cac0575e..eeeff2a04529 100644
--- a/include/linux/part_stat.h
+++ b/include/linux/part_stat.h
@@ -79,4 +79,6 @@ static inline void part_stat_set_all(struct block_device *part, int value)
 #define part_stat_local_read_cpu(part, field, cpu)			\
 	local_read(&(part_stat_get_cpu(part, field, cpu)))
 
+unsigned int bdev_count_inflight(struct block_device *part);
+
 #endif /* _LINUX_PART_STAT_H */
-- 
2.39.2