Messages in this thread | | | Subject | Re: [PATCH v2] block: fix alignment_offset math that assumes io_min is a power-of-2 | From | "Martin K. Petersen" <> | Date | Wed, 08 Oct 2014 20:23:10 -0400 |
| |
>>>>> "Mike" == Mike Snitzer <snitzer@redhat.com> writes:
Mike> The math in both blk_stack_limits() and Mike> queue_limit_alignment_offset() assume that a block device's io_min Mike> (aka minimum_io_size) is always a power-of-2. Fix the math such Mike> that it works for non-power-of-2 io_min.
Mike> This issue (of alignment_offset != 0) became apparent when testing Mike> dm-thinp with a thinp blocksize that matches a RAID6 stripesize of Mike> 1280K. Commit fdfb4c8c1 ("dm thin: set minimum_io_size to pool's Mike> data block size") unlocked the potential for alignment_offset != 0 Mike> due to the dm-thin-pool's io_min possibly being a non-power-of-2.
Looks good to me, Mike.
Acked-by: Martin K. Petersen <martin.petersen@oracle.com>
-- Martin K. Petersen Oracle Linux Engineering
| |