[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4bafa68f-46b3-7a76-de70-377a0f9cf130@opensource.wdc.com>
Date: Wed, 15 Jun 2022 20:56:23 +0900
From: Damien Le Moal <damien.lemoal@...nsource.wdc.com>
To: Pankaj Raghav <p.raghav@...sung.com>, hch@....de,
snitzer@...hat.com, axboe@...nel.dk
Cc: bvanassche@....org, linux-kernel@...r.kernel.org,
jiangbo.365@...edance.com, hare@...e.de, pankydev8@...il.com,
dm-devel@...hat.com, jonathan.derrick@...ux.dev,
gost.dev@...sung.com, dsterba@...e.com, jaegeuk@...nel.org,
linux-nvme@...ts.infradead.org, Johannes.Thumshirn@....com,
linux-block@...r.kernel.org,
Damien Le Moal <damien.lemoal@....com>,
Luis Chamberlain <mcgrof@...nel.org>
Subject: Re: [PATCH v7 06/13] null_blk: use zone_size_sects_shift for power of
2 zoned devices
On 6/15/22 19:19, Pankaj Raghav wrote:
> Instead of doing is_power_of_2 and ilog2 operation for every IO, cache
> the zone_size_sects_shift variable and use it for power of 2 zoned
> devices.
>
> This variable will be set to zero for non power of 2 zoned devices.
>
> Suggested-by: Damien Le Moal <damien.lemoal@....com>
> Reviewed-by: Hannes Reinecke <hare@...e.de>
> Reviewed-by: Luis Chamberlain <mcgrof@...nel.org>
> Signed-off-by: Pankaj Raghav <p.raghav@...sung.com>
> ---
> drivers/block/null_blk/null_blk.h | 6 ++++++
> drivers/block/null_blk/zoned.c | 11 ++++++++---
> 2 files changed, 14 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
> index 8359b4384..3bc7cbf25 100644
> --- a/drivers/block/null_blk/null_blk.h
> +++ b/drivers/block/null_blk/null_blk.h
> @@ -83,6 +83,12 @@ struct nullb_device {
> unsigned int imp_close_zone_no;
> struct nullb_zone *zones;
> sector_t zone_size_sects;
> + /*
> + * zone_size_sects_shift is only useful when the zone size is
> + * power of 2. This variable is set to zero when zone size is non
> + * power of 2.
> + */
> + unsigned int zone_size_sects_shift;
> bool need_zone_res_mgmt;
> spinlock_t zone_res_lock;
>
> diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
> index daf327015..5f929944b 100644
> --- a/drivers/block/null_blk/zoned.c
> +++ b/drivers/block/null_blk/zoned.c
> @@ -16,8 +16,8 @@ static inline sector_t mb_to_sects(unsigned long mb)
>
> static inline unsigned int null_zone_no(struct nullb_device *dev, sector_t sect)
> {
> - if (is_power_of_2(dev->zone_size_sects))
> - return sect >> ilog2(dev->zone_size_sects);
> + if (dev->zone_size_sects_shift)
> + return sect >> dev->zone_size_sects_shift;
>
> return div64_u64(sect, dev->zone_size_sects);
> }
> @@ -85,9 +85,14 @@ 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);
> +
> + if (is_power_of_2(dev->zone_size_sects))
> + dev->zone_size_sects_shift = ilog2(dev->zone_size_sects);
> + else
> + dev->zone_size_sects_shift = 0;
> +
> dev->nr_zones = DIV_ROUND_UP_SECTOR_T(dev_capacity_sects,
> dev->zone_size_sects);
> -
white line change.
This patch should be squashed with the previous one.
> dev->zones = kvmalloc_array(dev->nr_zones, sizeof(struct nullb_zone),
> GFP_KERNEL | __GFP_ZERO);
> if (!dev->zones)
--
Damien Le Moal
Western Digital Research
Powered by blists - more mailing lists