All of lore.kernel.org
 help / color / mirror / Atom feed
From: Pankaj Raghav <p.raghav@samsung.com>
To: hch@lst.de, snitzer@redhat.com, damien.lemoal@opensource.wdc.com,
	axboe@kernel.dk
Cc: bvanassche@acm.org, linux-kernel@vger.kernel.org,
	jiangbo.365@bytedance.com, hare@suse.de, pankydev8@gmail.com,
	dm-devel@redhat.com, jonathan.derrick@linux.dev,
	gost.dev@samsung.com, dsterba@suse.com, jaegeuk@kernel.org,
	linux-nvme@lists.infradead.org, Johannes.Thumshirn@wdc.com,
	linux-block@vger.kernel.org, Pankaj Raghav <p.raghav@samsung.com>,
	Luis Chamberlain <mcgrof@kernel.org>
Subject: [PATCH v7 05/13] null_blk: allow non power of 2 zoned devices
Date: Wed, 15 Jun 2022 12:19:12 +0200	[thread overview]
Message-ID: <20220615101920.329421-6-p.raghav@samsung.com> (raw)
In-Reply-To: <20220615101920.329421-1-p.raghav@samsung.com>

Convert the power of 2 based calculation with zone size to be generic in
null_zone_no with optimization for power of 2 based zone sizes.

The nr_zones calculation in null_init_zoned_dev has been replaced with a
division without special handling for power of 2 based zone sizes as
this function is called only during the initialization and will not
invoked in the hot path.

Performance Measurement:

Device:
zone size = 128M, blocksize=4k

FIO cmd:

fio --name=zbc --filename=/dev/nullb0 --direct=1 --zonemode=zbd  --size=23G
--io_size=<iosize> --ioengine=io_uring --iodepth=<iod> --rw=<mode> --bs=4k
--loops=4

The following results are an average of 4 runs on AMD Ryzen 5 5600X with
32GB of RAM:

Sequential Write:

x-----------------x---------------------------------x---------------------------------x
|     IOdepth     |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  578     |  2257    |   12.80   |  576     |  2248    |   25.78   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  581     |  2268    |   12.74   |  576     |  2248    |   25.85   |
x-----------------x---------------------------------x---------------------------------x

Sequential read:

x-----------------x---------------------------------x---------------------------------x
| IOdepth         |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  667     |  2605    |   11.79   |  675     |  2637    |   23.49   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  667     |  2605    |   11.79   |  675     |  2638    |   23.48   |
x-----------------x---------------------------------x---------------------------------x

Random read:

x-----------------x---------------------------------x---------------------------------x
| IOdepth         |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  522     |  2038    |   15.05   |  514     |  2006    |   30.87   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  522     |  2039    |   15.04   |  523     |  2042    |   30.33   |
x-----------------x---------------------------------x---------------------------------x

Minor variations are noticed in Sequential write with io depth 8 and
in random read with io depth 16. But overall no noticeable differences
were noticed

Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed by: Adam Manzanares <a.manzanares@samsung.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
---
 drivers/block/null_blk/main.c  |  5 ++---
 drivers/block/null_blk/zoned.c | 13 ++++++-------
 2 files changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index 6b67088f4..6c170927c 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -1929,9 +1929,8 @@ static int null_validate_conf(struct nullb_device *dev)
 	if (dev->queue_mode == NULL_Q_BIO)
 		dev->mbps = 0;
 
-	if (dev->zoned &&
-	    (!dev->zone_size || !is_power_of_2(dev->zone_size))) {
-		pr_err("zone_size must be power-of-two\n");
+	if (dev->zoned && !dev->zone_size) {
+		pr_err("Invalid zero zone size\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 2fdd7b20c..daf327015 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -16,7 +16,10 @@ static inline sector_t mb_to_sects(unsigned long mb)
 
 static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect)
 {
-	return sect >> ilog2(dev->zone_size_sects);
+	if (is_power_of_2(dev->zone_size_sects))
+		return sect >> ilog2(dev->zone_size_sects);
+
+	return div64_u64(sect, dev->zone_size_sects);
 }
 
 static inline void null_lock_zone_res(struct nullb_device *dev)
@@ -65,10 +68,6 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
 	sector_t sector = 0;
 	unsigned int i;
 
-	if (!is_power_of_2(dev->zone_size)) {
-		pr_err("zone_size must be power-of-two\n");
-		return -EINVAL;
-	}
 	if (dev->zone_size > dev->size) {
 		pr_err("Zone size larger than device capacity\n");
 		return -EINVAL;
@@ -86,8 +85,8 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
 	zone_capacity_sects = mb_to_sects(dev->zone_capacity);
 	dev_capacity_sects = mb_to_sects(dev->size);
 	dev->zone_size_sects = mb_to_sects(dev->zone_size);
-	dev->nr_zones = round_up(dev_capacity_sects, dev->zone_size_sects)
-		>> ilog2(dev->zone_size_sects);
+	dev->nr_zones =	DIV_ROUND_UP_SECTOR_T(dev_capacity_sects,
+					      dev->zone_size_sects);
 
 	dev->zones = kvmalloc_array(dev->nr_zones, sizeof(struct nullb_zone),
 				    GFP_KERNEL | __GFP_ZERO);
-- 
2.25.1


WARNING: multiple messages have this Message-ID (diff)
From: Pankaj Raghav <p.raghav@samsung.com>
To: hch@lst.de, snitzer@redhat.com, damien.lemoal@opensource.wdc.com,
	axboe@kernel.dk
Cc: Pankaj Raghav <p.raghav@samsung.com>,
	bvanassche@acm.org, pankydev8@gmail.com, gost.dev@samsung.com,
	jiangbo.365@bytedance.com, linux-nvme@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-block@vger.kernel.org,
	dm-devel@redhat.com, jonathan.derrick@linux.dev,
	Johannes.Thumshirn@wdc.com, dsterba@suse.com, jaegeuk@kernel.org,
	Luis Chamberlain <mcgrof@kernel.org>
Subject: [dm-devel] [PATCH v7 05/13] null_blk: allow non power of 2 zoned devices
Date: Wed, 15 Jun 2022 12:19:12 +0200	[thread overview]
Message-ID: <20220615101920.329421-6-p.raghav@samsung.com> (raw)
In-Reply-To: <20220615101920.329421-1-p.raghav@samsung.com>

Convert the power of 2 based calculation with zone size to be generic in
null_zone_no with optimization for power of 2 based zone sizes.

The nr_zones calculation in null_init_zoned_dev has been replaced with a
division without special handling for power of 2 based zone sizes as
this function is called only during the initialization and will not
invoked in the hot path.

Performance Measurement:

Device:
zone size = 128M, blocksize=4k

FIO cmd:

fio --name=zbc --filename=/dev/nullb0 --direct=1 --zonemode=zbd  --size=23G
--io_size=<iosize> --ioengine=io_uring --iodepth=<iod> --rw=<mode> --bs=4k
--loops=4

The following results are an average of 4 runs on AMD Ryzen 5 5600X with
32GB of RAM:

Sequential Write:

x-----------------x---------------------------------x---------------------------------x
|     IOdepth     |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  578     |  2257    |   12.80   |  576     |  2248    |   25.78   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  581     |  2268    |   12.74   |  576     |  2248    |   25.85   |
x-----------------x---------------------------------x---------------------------------x

Sequential read:

x-----------------x---------------------------------x---------------------------------x
| IOdepth         |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  667     |  2605    |   11.79   |  675     |  2637    |   23.49   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  667     |  2605    |   11.79   |  675     |  2638    |   23.48   |
x-----------------x---------------------------------x---------------------------------x

Random read:

x-----------------x---------------------------------x---------------------------------x
| IOdepth         |            8                    |            16                   |
x-----------------x---------------------------------x---------------------------------x
|                 |  KIOPS   |BW(MiB/s) | Lat(usec) |  KIOPS   |BW(MiB/s) | Lat(usec) |
x-----------------x---------------------------------x---------------------------------x
| Without patch   |  522     |  2038    |   15.05   |  514     |  2006    |   30.87   |
x-----------------x---------------------------------x---------------------------------x
|  With patch     |  522     |  2039    |   15.04   |  523     |  2042    |   30.33   |
x-----------------x---------------------------------x---------------------------------x

Minor variations are noticed in Sequential write with io depth 8 and
in random read with io depth 16. But overall no noticeable differences
were noticed

Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed by: Adam Manzanares <a.manzanares@samsung.com>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
---
 drivers/block/null_blk/main.c  |  5 ++---
 drivers/block/null_blk/zoned.c | 13 ++++++-------
 2 files changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index 6b67088f4..6c170927c 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -1929,9 +1929,8 @@ static int null_validate_conf(struct nullb_device *dev)
 	if (dev->queue_mode == NULL_Q_BIO)
 		dev->mbps = 0;
 
-	if (dev->zoned &&
-	    (!dev->zone_size || !is_power_of_2(dev->zone_size))) {
-		pr_err("zone_size must be power-of-two\n");
+	if (dev->zoned && !dev->zone_size) {
+		pr_err("Invalid zero zone size\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 2fdd7b20c..daf327015 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -16,7 +16,10 @@ static inline sector_t mb_to_sects(unsigned long mb)
 
 static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect)
 {
-	return sect >> ilog2(dev->zone_size_sects);
+	if (is_power_of_2(dev->zone_size_sects))
+		return sect >> ilog2(dev->zone_size_sects);
+
+	return div64_u64(sect, dev->zone_size_sects);
 }
 
 static inline void null_lock_zone_res(struct nullb_device *dev)
@@ -65,10 +68,6 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
 	sector_t sector = 0;
 	unsigned int i;
 
-	if (!is_power_of_2(dev->zone_size)) {
-		pr_err("zone_size must be power-of-two\n");
-		return -EINVAL;
-	}
 	if (dev->zone_size > dev->size) {
 		pr_err("Zone size larger than device capacity\n");
 		return -EINVAL;
@@ -86,8 +85,8 @@ int null_init_zoned_dev(struct nullb_device *dev, struct request_queue *q)
 	zone_capacity_sects = mb_to_sects(dev->zone_capacity);
 	dev_capacity_sects = mb_to_sects(dev->size);
 	dev->zone_size_sects = mb_to_sects(dev->zone_size);
-	dev->nr_zones = round_up(dev_capacity_sects, dev->zone_size_sects)
-		>> ilog2(dev->zone_size_sects);
+	dev->nr_zones =	DIV_ROUND_UP_SECTOR_T(dev_capacity_sects,
+					      dev->zone_size_sects);
 
 	dev->zones = kvmalloc_array(dev->nr_zones, sizeof(struct nullb_zone),
 				    GFP_KERNEL | __GFP_ZERO);
-- 
2.25.1

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel


  parent reply	other threads:[~2022-06-15 10:20 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20220615101924eucas1p27fbce623c0e1b3097169bf23dd6266d8@eucas1p2.samsung.com>
2022-06-15 10:19 ` [PATCH v7 00/13] support non power of 2 zoned device Pankaj Raghav
2022-06-15 10:19   ` [dm-devel] " Pankaj Raghav
     [not found]   ` <CGME20220615101927eucas1p17220c7a36f69f59ff8ddd560b42967ec@eucas1p1.samsung.com>
2022-06-15 10:19     ` [PATCH v7 01/13] block: make blkdev_nr_zones and blk_queue_zone_no generic for npo2 zsze Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
2022-06-15 20:18       ` Bart Van Assche
2022-06-15 20:18         ` Bart Van Assche
     [not found]   ` <CGME20220615101931eucas1p15ed09ae433a2c378b599e9086130d8eb@eucas1p1.samsung.com>
2022-06-15 10:19     ` [PATCH v7 02/13] block: allow blk-zoned devices to have non-power-of-2 zone size Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
2022-06-15 20:28       ` Bart Van Assche
2022-06-15 20:28         ` Bart Van Assche
2022-06-16 10:09         ` Pankaj Raghav
2022-06-16 10:09           ` Pankaj Raghav
2022-06-16 16:04           ` Luis Chamberlain
2022-06-16 16:04             ` Luis Chamberlain
2022-06-16 23:30           ` Damien Le Moal
2022-06-16 23:30             ` Damien Le Moal
     [not found]   ` <CGME20220615101935eucas1p26a7bc245d88a89312158d7a265f64aef@eucas1p2.samsung.com>
2022-06-15 10:19     ` [PATCH v7 03/13] nvme: zns: Allow ZNS drives that have non-power_of_2 " Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
     [not found]   ` <CGME20220615101938eucas1p26ab159a1ffd0fa5a16d7f202ba7206e7@eucas1p2.samsung.com>
2022-06-15 10:19     ` [PATCH v7 04/13] nvmet: Allow ZNS target to support non-power_of_2 zone sizes Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
     [not found]   ` <CGME20220615101941eucas1p25e1c27b363e6b288b848521298e31705@eucas1p2.samsung.com>
2022-06-15 10:19     ` Pankaj Raghav [this message]
2022-06-15 10:19       ` [dm-devel] [PATCH v7 05/13] null_blk: allow non power of 2 zoned devices Pankaj Raghav
     [not found]   ` <CGME20220615101945eucas1p16fa264e81d9b6027ff131dd311ed91e2@eucas1p1.samsung.com>
2022-06-15 10:19     ` [PATCH v7 06/13] null_blk: use zone_size_sects_shift for " Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
2022-06-15 11:56       ` Damien Le Moal
2022-06-15 11:56         ` [dm-devel] " Damien Le Moal
     [not found]   ` <CGME20220615101948eucas1p2d8d801735c39b25256a019134adb0c6f@eucas1p2.samsung.com>
2022-06-15 10:19     ` [PATCH v7 07/13] zonefs: allow non " Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
     [not found]   ` <CGME20220615101951eucas1p238eb45e563bd9645af81bf16c56d98ec@eucas1p2.samsung.com>
2022-06-15 10:19     ` [PATCH v7 08/13] dm-zoned: ensure only power of 2 zone sizes are allowed Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
     [not found]   ` <CGME20220615101955eucas1p19b9d42ead7331f69f7dad1ec100312c2@eucas1p1.samsung.com>
2022-06-15 10:19     ` [dm-devel] [PATCH v7 09/13] dm-zone: use generic helpers to calculate offset from zone start Pankaj Raghav
2022-06-15 10:19       ` Pankaj Raghav
     [not found]   ` <CGME20220615102000eucas1p27720aaa3c309327b2b9a33c5f840f498@eucas1p2.samsung.com>
2022-06-15 10:19     ` [dm-devel] [PATCH v7 10/13] dm-table: use bdev_is_zone_start helper in device_area_is_invalid() Pankaj Raghav
2022-06-15 10:19       ` Pankaj Raghav
2022-06-15 11:53       ` Damien Le Moal
2022-06-15 11:53         ` [dm-devel] " Damien Le Moal
2022-06-16  9:55         ` Pankaj Raghav
2022-06-16  9:55           ` [dm-devel] " Pankaj Raghav
2022-06-16 23:29           ` Damien Le Moal
2022-06-16 23:29             ` [dm-devel] " Damien Le Moal
     [not found]   ` <CGME20220615102004eucas1p1e458ea097d381058b16fc6daa3eec998@eucas1p1.samsung.com>
2022-06-15 10:19     ` [dm-devel] [PATCH v7 11/13] dm-table: allow non po2 zoned devices Pankaj Raghav
2022-06-15 10:19       ` Pankaj Raghav
     [not found]   ` <CGME20220615102007eucas1p1106f9520e2a86beb3792107dffd8071b@eucas1p1.samsung.com>
2022-06-15 10:19     ` [dm-devel] [PATCH v7 12/13] dm: call dm_zone_endio after the target endio callback for " Pankaj Raghav
2022-06-15 10:19       ` Pankaj Raghav
2022-06-15 11:01       ` [dm-devel] " Damien Le Moal
2022-06-15 11:01         ` Damien Le Moal
2022-06-16 12:24         ` Pankaj Raghav
2022-06-16 12:24           ` Pankaj Raghav
2022-06-16 23:33           ` Damien Le Moal
2022-06-16 23:33             ` Damien Le Moal
     [not found]   ` <CGME20220615102011eucas1p220368db4a186181b1927dea50a79e5d4@eucas1p2.samsung.com>
2022-06-15 10:19     ` [PATCH v7 13/13] dm: add non power of 2 zoned target Pankaj Raghav
2022-06-15 10:19       ` [dm-devel] " Pankaj Raghav
2022-06-15 11:49       ` Damien Le Moal
2022-06-15 11:49         ` [dm-devel] " Damien Le Moal
2022-06-16 16:12         ` Pankaj Raghav
2022-06-16 16:12           ` [dm-devel] " Pankaj Raghav
2022-06-16 23:49           ` Damien Le Moal
2022-06-16 23:49             ` [dm-devel] " Damien Le Moal
2022-06-17  5:45             ` Pankaj Raghav
2022-06-17  5:45               ` [dm-devel] " Pankaj Raghav
2022-06-17  6:12               ` Damien Le Moal
2022-06-17  6:12                 ` [dm-devel] " Damien Le Moal
2022-06-17  6:40                 ` Pankaj Raghav
2022-06-17  6:40                   ` [dm-devel] " Pankaj Raghav
2022-06-17  6:56                   ` Damien Le Moal
2022-06-17  6:56                     ` [dm-devel] " Damien Le Moal
2022-06-17  8:03                     ` Pankaj Raghav
2022-06-17  8:03                       ` [dm-devel] " Pankaj Raghav
2022-06-15 14:19       ` kernel test robot
2022-06-15 14:19         ` [dm-devel] " kernel test robot
2022-06-15 19:54       ` Randy Dunlap
2022-06-15 19:54         ` [dm-devel] " Randy Dunlap
2022-06-16 10:28         ` Pankaj Raghav
2022-06-16 10:28           ` [dm-devel] " Pankaj Raghav

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220615101920.329421-6-p.raghav@samsung.com \
    --to=p.raghav@samsung.com \
    --cc=Johannes.Thumshirn@wdc.com \
    --cc=axboe@kernel.dk \
    --cc=bvanassche@acm.org \
    --cc=damien.lemoal@opensource.wdc.com \
    --cc=dm-devel@redhat.com \
    --cc=dsterba@suse.com \
    --cc=gost.dev@samsung.com \
    --cc=hare@suse.de \
    --cc=hch@lst.de \
    --cc=jaegeuk@kernel.org \
    --cc=jiangbo.365@bytedance.com \
    --cc=jonathan.derrick@linux.dev \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=mcgrof@kernel.org \
    --cc=pankydev8@gmail.com \
    --cc=snitzer@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.