On Thu, Jul 24, 2014 at 5:25 PM, Alex Elder <elder@xxxxxxxx> wrote:
> On 07/24/2014 03:42 AM, Ilya Dryomov wrote:
>> There is no sense in trying to update the mapping size before it's even
>> been set.
>
> It took me a bit to follow this. But basically there is
> no mapping unless it's mapped. So previously this was
> updating the mapping information even for unmapped
> parent (or other) images. There's no need to update
> the mapping size for a snapshot--it'll never change.
>
> Is that right? If not, please advise; otherwise:
No. Previously it was updating the mapping size both on the inital map
and on refresh (of the base image). Doing that on the inital map
doesn't make sense: not only struct rbd_mapping fields aren't properly
initialized at that point - rbd_dev_mapping_set() is called much later
in the map sequence, snap_id isn't initialized either (at least in the
format 2 case, I haven't looked too closely at the format 1 case). And
just in general, trying to update something before it had a chance to
change is bogus..
Thanks,
Ilya
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
