[dm-devel] Re: [PATCH v2] dm: add topology support

[Mike Snitzer]

On Wed, Jun 10 2009 at  2:51am -0400,
Martin K. Petersen <martin.petersen at oracle.com> wrote:

> >>>>> "Mike" == Mike Snitzer <snitzer at redhat.com> writes:
> 
> Mike> Use blk_stack_limits() to stack block limits (including topology)
> Mike> rather than duplicate the equivalent within Device Mapper.
> 
> Great!  So how are the userland bits coming along?

They should be pretty well sorted out with the patches I posted to
lvm-devel early Monday morning (code review still needed):
https://www.redhat.com/archives/lvm-devel/2009-June/msg00037.html

As for the DM changes.  Alasdair reviewed v2 of the patch and found an
issue that I need to get a final answer to.

Your change to dm-table.c:dm_table_set_restrictions() in linux-2.6-block
+for-next's ae03bf639a5027d27270123f5f6e3ee6a412781d introduced calls to
blk_queue_*() setters.

My v2 of the DM topology support patch does away with those and just
uses blk_stack_limits().  In the process we go back to _not_ using the
blk_queue_*() setters (note the additional checks that those setters
have).

So the question: is _not_ using the blk_queue_*() setters perfectly
fine?  Given that DM has always _not_ used them the quick answer is
"seems fine".

But I need to dig a bit more to understand if the additional logic in
the blk_queue_*() setters is something DM shouldn't be circumventing.

But we're almost done with these DM/LVM2 topology bits.. really :)

Mike