On 11/1/24 14:21, Christoph Hellwig wrote:
> Otherwise it can create unaligned writes on zoned devices.
>
> Fixes: a805a4fa4fa3 ("block: introduce zone_write_granularity limit")
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> ---
> block/blk-merge.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/block/blk-merge.c b/block/blk-merge.c
> index d813d799cee7..d6895859a2fb 100644
> --- a/block/blk-merge.c
> +++ b/block/blk-merge.c
> @@ -358,7 +358,8 @@ int bio_split_rw_at(struct bio *bio, const struct queue_limits *lim,
> * split size so that each bio is properly block size aligned, even if
> * we do not use the full hardware limits.
> */
> - bytes = ALIGN_DOWN(bytes, lim->logical_block_size);
> + bytes = ALIGN_DOWN(bytes, lim->zone_write_granularity ?
> + lim->zone_write_granularity : lim->logical_block_size);
Nit: we could also do:
bytes = ALIGN_DOWN(bytes,
max(lim->logical_block_size, lim->zone_write_granularity));
Also, I wonder if we should leave read split as is based on the logical block
size only ? Probably does not matter much...
>
> /*
> * Bio splitting may cause subtle trouble such as hang when doing sync
--
Damien Le Moal
Western Digital Research
