null_blk: Do not allow runt zone with zone capacity smaller then zone size

A zoned device with a smaller last zone together with a zone capacity
smaller than the zone size does make any sense as that does not
correspond to any possible setup for a real device:
1) For ZNS and zoned UFS devices, all zones are always the same size.
2) For SMR HDDs, all zones always have the same capacity.
In other words, if we have a smaller last runt zone, then this zone
capacity should always be equal to the zone size.

Add a check in null_init_zoned_dev() to prevent a configuration to have
both a smaller zone size and a zone capacity smaller than the zone size.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Niklas Cassel <cassel@kernel.org>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20240530054035.491497-2-dlemoal@kernel.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>

diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 79c8e5e..f118d30 100644 (file)
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -74,6 +74,17 @@ int null_init_zoned_dev(struct nullb_device *dev,
                return -EINVAL;
        }
 
+       /*
+        * If a smaller zone capacity was requested, do not allow a smaller last
+        * zone at the same time as such zone configuration does not correspond
+        * to any real zoned device.
+        */
+       if (dev->zone_capacity != dev->zone_size &&
+           dev->size & (dev->zone_size - 1)) {
+               pr_err("A smaller last zone is not allowed with zone capacity smaller than zone size.\n");
+               return -EINVAL;
+       }
+
        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);