Before if the block layer's limit stacking didn't establish an optimal_io_size that was compatible with the dm-thin-pool's data block size of the we'd set optimal_io_size to the data block size and minimum_io_size to 0 (which the block layer adjusts to be physical_block_size). Update pool_io_hints() to set both minimum_io_size and optimal_io_size to the pool's data block size. This fixes an issue reported where mkfs.xfs would create more XFS Allocation Groups on thinp volumes than on a normal linear LV of comparable size, see: https://bugzilla.redhat.com/show_bug.cgi?id=1003227 Reported-by: Chris Murphy <lists@xxxxxxxxxxxxxxxxx> Signed-off-by: Mike Snitzer <snitzer@xxxxxxxxxx> --- drivers/md/dm-thin.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c index fc9c848..c92e1b9 100644 --- a/drivers/md/dm-thin.c +++ b/drivers/md/dm-thin.c @@ -3112,7 +3112,7 @@ static void pool_io_hints(struct dm_target *ti, struct queue_limits *limits) */ if (io_opt_sectors < pool->sectors_per_block || do_div(io_opt_sectors, pool->sectors_per_block)) { - blk_limits_io_min(limits, 0); + blk_limits_io_min(limits, pool->sectors_per_block << SECTOR_SHIFT); blk_limits_io_opt(limits, pool->sectors_per_block << SECTOR_SHIFT); } -- 1.7.4.4 -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
