[PATCH 04/16] block: allow blk-zoned devices to have non-power-of-2 zone size

[Damien Le Moal]

On 4/28/22 01:02, Pankaj Raghav wrote:
> Convert the calculations on zone size to be generic instead of relying on
> power_of_2 based logic in the block layer using the helpers wherever
> possible.
> 
> The only hot path affected by this change for power_of_2 zoned devices
> is in blk_check_zone_append() but the effects should be negligible as the
> helper blk_queue_zone_aligned() optimizes the calculation for those
> devices. Note that the append path cannot be accessed by direct raw access
> to the block device but only through a filesystem abstraction.
> 
> Finally, remove the check for power_of_2 zone size requirement in
> blk-zoned.c
> 
> Reviewed-by: Luis Chamberlain <mcgrof at kernel.org>
> Signed-off-by: Pankaj Raghav <p.raghav at samsung.com>
> ---
>  block/blk-core.c  |  3 +--
>  block/blk-zoned.c | 12 ++++++------
>  2 files changed, 7 insertions(+), 8 deletions(-)
> 
> diff --git a/block/blk-core.c b/block/blk-core.c
> index 937bb6b86331..850caf311064 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -634,8 +634,7 @@ static inline blk_status_t blk_check_zone_append(struct request_queue *q,
>  		return BLK_STS_NOTSUPP;
>  
>  	/* The bio sector must point to the start of a sequential zone */
> -	if (pos & (blk_queue_zone_sectors(q) - 1) ||
> -	    !blk_queue_zone_is_seq(q, pos))
> +	if (!blk_queue_zone_aligned(q, pos) || !blk_queue_zone_is_seq(q, pos))

blk_queue_zone_aligned() is a little confusing since "aligned" is also
used for write-pointer aligned. I would rename this helper

blk_queue_is_zone_start()

or something like that.


>  		return BLK_STS_IOERR;
>  
>  	/*
> diff --git a/block/blk-zoned.c b/block/blk-zoned.c
> index 1dff4a8bd51d..f7c7c3bd148d 100644
> --- a/block/blk-zoned.c
> +++ b/block/blk-zoned.c
> @@ -288,10 +288,10 @@ int blkdev_zone_mgmt(struct block_device *bdev, enum req_opf op,
>  		return -EINVAL;
>  
>  	/* Check alignment (handle eventual smaller last zone) */
> -	if (sector & (zone_sectors - 1))
> +	if (!blk_queue_zone_aligned(q, sector))
>  		return -EINVAL;
>  
> -	if ((nr_sectors & (zone_sectors - 1)) && end_sector != capacity)
> +	if (!blk_queue_zone_aligned(q, nr_sectors) && end_sector != capacity)
>  		return -EINVAL;
>  
>  	/*
> @@ -489,14 +489,14 @@ static int blk_revalidate_zone_cb(struct blk_zone *zone, unsigned int idx,
>  	 * smaller last zone.
>  	 */
>  	if (zone->start == 0) {
> -		if (zone->len == 0 || !is_power_of_2(zone->len)) {
> -			pr_warn("%s: Invalid zoned device with non power of two zone size (%llu)\n",
> -				disk->disk_name, zone->len);
> +		if (zone->len == 0) {
> +			pr_warn("%s: Invalid zoned device size",
> +				disk->disk_name);

The message is weird now. Please change it to "Invalid zone size".

Also, the entire premise of this patch series is that it is hard for
people to support the unusable sectors between zone capacity and zone end
for drives with a zone capacity smaller than the zone size.

Yet, here you do not check that zone capacity == zone size for drives that
do not have a zone size equal to a power of 2 number of sectors. This
means that we can still have drives with ZC < ZS AND ZS not equal to a
power of 2. So from the point of view of your arguments, no gains at all.
Any thoughts on this ?

>  			return -ENODEV;
>  		}
>  
>  		args->zone_sectors = zone->len;
> -		args->nr_zones = (capacity + zone->len - 1) >> ilog2(zone->len);
> +		args->nr_zones = div64_u64(capacity + zone->len - 1, zone->len);
>  	} else if (zone->start + args->zone_sectors < capacity) {
>  		if (zone->len != args->zone_sectors) {
>  			pr_warn("%s: Invalid zoned device with non constant zone size\n",


-- 
Damien Le Moal
Western Digital Research