Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Adam Manzanares <a.manzanares@samsung.com>
To:     Pankaj Raghav <p.raghav@samsung.com>
CC:     "jaegeuk@kernel.org" <jaegeuk@kernel.org>,
        "axboe@kernel.dk" <axboe@kernel.dk>,
        "snitzer@kernel.org" <snitzer@kernel.org>,
        "hch@lst.de" <hch@lst.de>, "mcgrof@kernel.org" <mcgrof@kernel.org>,
        "naohiro.aota@wdc.com" <naohiro.aota@wdc.com>,
        "sagi@grimberg.me" <sagi@grimberg.me>,
        "damien.lemoal@opensource.wdc.com" <damien.lemoal@opensource.wdc.com>,
        "dsterba@suse.com" <dsterba@suse.com>,
        "johannes.thumshirn@wdc.com" <johannes.thumshirn@wdc.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-btrfs@vger.kernel.org" <linux-btrfs@vger.kernel.org>,
        "clm@fb.com" <clm@fb.com>,
        "gost.dev@samsung.com" <gost.dev@samsung.com>,
        "chao@kernel.org" <chao@kernel.org>,
        "linux-f2fs-devel@lists.sourceforge.net" 
        <linux-f2fs-devel@lists.sourceforge.net>,
        "josef@toxicpanda.com" <josef@toxicpanda.com>,
        "jonathan.derrick@linux.dev" <jonathan.derrick@linux.dev>,
        "agk@redhat.com" <agk@redhat.com>,
        "kbusch@kernel.org" <kbusch@kernel.org>,
        "kch@nvidia.com" <kch@nvidia.com>,
        "linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
        "dm-devel@redhat.com" <dm-devel@redhat.com>,
        "bvanassche@acm.org" <bvanassche@acm.org>,
        "jiangbo.365@bytedance.com" <jiangbo.365@bytedance.com>,
        "linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
        "matias.bjorling@wdc.com" <matias.bjorling@wdc.com>,
        "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>
Subject: Re: [PATCH 13/16] null_blk: allow non power of 2 zoned devices
Thread-Topic: [PATCH 13/16] null_blk: allow non power of 2 zoned devices
Thread-Index: AQHYWlBN4zB7Wf7sW0aEyMwWol2CyK0HnkAA
Date:   Fri, 29 Apr 2022 17:30:07 +0000
Message-ID: <20220429173006.GD174938@bgt-140510-bm01>
In-Reply-To: <20220427160255.300418-14-p.raghav@samsung.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-ID: <B4D2F0A8C589E242945A08E1571AAD23@ssi.samsung.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
CMS-TYPE: 301P
References: <20220427160255.300418-1-p.raghav@samsung.com>
        <CGME20220427160310eucas1p28cd3c5ff4fb7a04bc77c4c0b9d96bb74@eucas1p2.samsung.com>
        <20220427160255.300418-14-p.raghav@samsung.com>
Precedence: bulk

On Wed, Apr 27, 2022 at 06:02:52PM +0200, Pankaj Raghav wrote:
> 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.
>=20
> 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.
>=20
> Performance Measurement:
>=20
> Device:
> zone size =3D 128M, blocksize=3D4k
>=20
> FIO cmd:
>=20
> fio --name=3Dzbc --filename=3D/dev/nullb0 --direct=3D1 --zonemode=3Dzbd  =
--size=3D23G
> --io_size=3D<iosize> --ioengine=3Dio_uring --iodepth=3D<iod> --rw=3D<mode=
> --bs=3D4k
> --loops=3D4
>=20
> The following results are an average of 4 runs on AMD Ryzen 5 5600X with
> 32GB of RAM:
>=20
> Sequential Write:
>=20
> 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
>=20
> Sequential read:
>=20
> 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
>=20
> Random read:
>=20
> 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
>=20
> 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
>=20
> Reviewed-by: Luis Chamberlain <mcgrof@kernel.org>
> Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
> ---
>  drivers/block/null_blk/main.c  |  4 ++--
>  drivers/block/null_blk/zoned.c | 14 +++++++-------
>  2 files changed, 9 insertions(+), 9 deletions(-)
>=20
> diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.=
c
> index c441a4972064..82a62b543782 100644
> --- a/drivers/block/null_blk/main.c
> +++ b/drivers/block/null_blk/main.c
> @@ -1931,8 +1931,8 @@ static int null_validate_conf(struct nullb_device *=
dev)
>  		dev->mbps =3D 0;
> =20
>  	if (dev->zoned &&
> -	    (!dev->zone_size || !is_power_of_2(dev->zone_size))) {
> -		pr_err("zone_size must be power-of-two\n");
> +	    (!dev->zone_size)) {
> +		pr_err("zone_size must not be zero\n");
>  		return -EINVAL;
>  	}
> =20
> diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zone=
d.c
> index dae54dd1aeac..00c34e65ef0a 100644
> --- a/drivers/block/null_blk/zoned.c
> +++ b/drivers/block/null_blk/zoned.c
> @@ -13,7 +13,10 @@ static inline sector_t mb_to_sects(unsigned long mb)
> =20
>  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);
>  }
> =20
>  static inline void null_lock_zone_res(struct nullb_device *dev)
> @@ -62,10 +65,6 @@ int null_init_zoned_dev(struct nullb_device *dev, stru=
ct request_queue *q)
>  	sector_t sector =3D 0;
>  	unsigned int i;
> =20
> -	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;
> @@ -83,8 +82,9 @@ int null_init_zoned_dev(struct nullb_device *dev, struc=
t request_queue *q)
>  	zone_capacity_sects =3D mb_to_sects(dev->zone_capacity);
>  	dev_capacity_sects =3D mb_to_sects(dev->size);
>  	dev->zone_size_sects =3D mb_to_sects(dev->zone_size);
> -	dev->nr_zones =3D round_up(dev_capacity_sects, dev->zone_size_sects)
> -		>> ilog2(dev->zone_size_sects);
> +	dev->nr_zones =3D
> +		div64_u64(roundup(dev_capacity_sects, dev->zone_size_sects),
> +			  dev->zone_size_sects);
> =20
>  	dev->zones =3D kvmalloc_array(dev->nr_zones, sizeof(struct nullb_zone),
>  				    GFP_KERNEL | __GFP_ZERO);
> --=20
> 2.25.1
>


Looks good.

Reviewed by: Adam Manzanares <a.manzanares@samsung.com>=